{
  "version": 3,
  "sources": ["ssg:https://framerusercontent.com/modules/rgSgXwaVGWG36Z82z6xf/5VuqIJ2aK97jABpIQNLn/a3ufc1L7f-2.js"],
  "sourcesContent": ["import{jsx as e,jsxs as t}from\"react/jsx-runtime\";import{ComponentPresetsConsumer as a,Link as i}from\"framer\";import{motion as n}from\"framer-motion\";import*as o from\"react\";import{Youtube as r}from\"https://framerusercontent.com/modules/NEd4VmDdsxM3StIUbddO/bZxrMUxBPAhoXlARkK9C/YouTube.js\";export const richText=/*#__PURE__*/t(o.Fragment,{children:[/*#__PURE__*/e(\"p\",{children:\"Businesses increasingly turn to AI technologies - from Large Language Models (LLMs) to computer vision systems - to transform their operations. Typical applications include process automation, advanced data analytics, and personalised customer experiences. Almost every industry will benefit from what AI has to offer. However, successful implementation requires careful consideration of the technology and the organisation's data strategy, technical capabilities, and long-term objectives. However, many business leaders ask if they should build their own AI software from scratch or buy an off-the-shelf solution. Each option has advantages and challenges, and the right choice depends on the organisation\u2019s needs, technical capabilities, and business goals.\"}),/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"Customisation\\xa0\"})}),/*#__PURE__*/e(\"p\",{children:\"One of the most significant distinctions between building and buying AI-powered software is the level of customisation needed. Sometimes, off-the-shelf software won\u2019t cut it, and you need a custom solution that sets you apart or caters to your unique needs. Creating AI software in-house allows you to tailor every feature to align precisely with your company\u2019s unique business objectives. For example, in the live entertainment space, we\u2019ve seen companies interested in the level of individualisation that AI offers by building unique custom fan experiences to deepen their audience's connection to their events, sports teams and artists. This type of AI-powered software is designed and engineered to meet the specific needs of the fans on their favourite devices. For instance, one entertainment client leveraged AI to analyse live sports feeds to pull highlights out in real time and serve them on an individualised basis, meeting fans' exact tastes.\"}),/*#__PURE__*/e(\"p\",{children:\"Off-the-shelf AI solutions often cater to common use cases like customer service chatbots or predictive analytics. While these solutions excel at standardised tasks like sentiment analysis or basic forecasting, they may need to improve when dealing with industry-specific challenges or unique data sets. Buying might make more sense if your needs align well with these standard functionalities and you\u2019re looking to implement a solution immediately.\\xa0\"}),/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"Cost\"})}),/*#__PURE__*/e(\"p\",{children:\"Cost is another major factor to take into consideration. Building custom AI-powered (or any software in general) can carry significant expenses. You\u2019ll need to hire AI architects, data scientists, AI engineers, product designers and product managers to do this right. Initial development costs typically range from $200,000 to $1.5+ million depending on complexity, with ongoing development averaging 15-20% of the initial investment annually. You can do this in-house if you already have these roles or hire outside talent on a project basis. Then, you must also remember the ongoing cost of maintenance, upgrades, and updating as data changes.\"}),/*#__PURE__*/e(\"p\",{children:\"By contrast, buying an off-the-shelf solution reduces these up-front costs. Many AI SaaS companies offer subscription models which allow for predictable budgeting. However, in the end, it may only partially do what you need because it was built to serve the masses. While a SaaS solution might cost $50,000-$150,000 annually, scaling costs can increase dramatically with user count or API calls, potentially exceeding custom development costs over a 3-5-year period.\"}),/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"Data Ownership\\xa0\\xa0\"})}),/*#__PURE__*/e(\"p\",{children:\"Building AI-powered software internally gives you greater control over data handling and model customisation, which is especially important in industries with strict data privacy regulations. You own the intellectual property and can adjust the model as needs evolve, which is a significant advantage in highly regulated fields like finance and healthcare. This becomes particularly crucial when dealing with sensitive data like HIPAA-protected health records or GDPR-regulated personal information, where data lineage and processing transparency are mandatory.\"}),/*#__PURE__*/e(\"p\",{children:\"When buying off-the-shelf solutions, control over data usage, privacy, and updates lies primarily with the SaaS company. They may use aggregated data from all clients to improve their models, which could challenge data security. Additionally, vendor lock-in becomes a genuine concern - migrating your data and retraining models with a new vendor can be costly and time-consuming.\"}),/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"Implementation Speed\"})}),/*#__PURE__*/e(\"p\",{children:\"While traditional AI development cycles could take 12-18 months, modern AI-enhanced development practices have dramatically reduced this timeline. AI-integrated SDLC leverages:\"}),/*#__PURE__*/t(\"ul\",{children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",style:{\"--framer-font-size\":\"11px\",\"--framer-text-color\":\"rgb(0, 0, 0)\",\"--framer-text-decoration\":\"none\"},children:/*#__PURE__*/e(\"p\",{children:\"AI-assisted code generation and testing\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",style:{\"--framer-font-size\":\"11px\",\"--framer-text-color\":\"rgb(0, 0, 0)\",\"--framer-text-decoration\":\"none\"},children:/*#__PURE__*/e(\"p\",{children:\"AI-assisted code review and optimisation\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",style:{\"--framer-font-size\":\"11px\",\"--framer-text-color\":\"rgb(0, 0, 0)\",\"--framer-text-decoration\":\"none\"},children:/*#__PURE__*/e(\"p\",{children:\"Continuous model evaluation\"})})]}),/*#__PURE__*/e(\"p\",{children:\"For example, engineers are now using AI tools to supercharge their workflows by automating repetitive tasks so they can focus on writing new features. Product managers and product owners are also using AI to develop well-defined user stories that are force-ranked and constantly reprioritised to ensure that the engineers are spending their valuable time on the most necessary features - and that they are building those features precisely as the product owner intended. If your business seeks a unique solution to gain a competitive edge, investing in a build is worthwhile and may be quicker than anticipated.\\xa0\"}),/*#__PURE__*/e(\"p\",{children:\"When buying off-the-shelf AI, many of these solutions come with pre-trained models, which deliver value almost immediately if you want to standardise processes or enhance productivity where differentiation isn\u2019t critical. However, be prepared for a 3-6 month integration period to properly configure, test, and train staff on the new system.\"}),/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"Long-Term Scalability\\xa0\"})}),/*#__PURE__*/e(\"p\",{children:\"Custom AI-powered software offers long-term flexibility, as your team can refine or scale the system as your business grows, adapting to changing industry trends and new data inputs. This includes the ability to:\"}),/*#__PURE__*/t(\"ul\",{children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",style:{\"--framer-font-size\":\"11px\",\"--framer-text-color\":\"rgb(0, 0, 0)\",\"--framer-text-decoration\":\"none\"},children:/*#__PURE__*/e(\"p\",{children:\"Switch between different AI models as technology evolves\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",style:{\"--framer-font-size\":\"11px\",\"--framer-text-color\":\"rgb(0, 0, 0)\",\"--framer-text-decoration\":\"none\"},children:/*#__PURE__*/e(\"p\",{children:\"Scale horizontally across multiple cloud providers\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",style:{\"--framer-font-size\":\"11px\",\"--framer-text-color\":\"rgb(0, 0, 0)\",\"--framer-text-decoration\":\"none\"},children:/*#__PURE__*/e(\"p\",{children:\"Integrate new data sources and features without architectural overhaul\"})})]}),/*#__PURE__*/e(\"p\",{children:\"Buying an AI solution will only be future-proof if the vendor invests in improving their products and is aligned with your evolving needs. So, in this case, you are taking a big chance on the vendor and should ensure long-term scalability, which is important to them before investing.\\xa0\"}),/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"Decision Framework\\xa0\"})}),/*#__PURE__*/e(\"p\",{children:\"Before making your choice, consider these key factors:\"}),/*#__PURE__*/t(\"ol\",{children:[/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Data Uniqueness: How specific is your data to your industry/business?\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Competitive Advantage: Will AI differentiation drive business value?\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Technical Capability: Can you support ongoing AI development?\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Time to Market: What's your implementation deadline?\"})}),/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",children:/*#__PURE__*/e(\"p\",{children:\"Budget Structure: Do you prefer CapEx or OpEx investment?\"})})]}),/*#__PURE__*/t(\"h6\",{children:[/*#__PURE__*/e(\"strong\",{children:\"Have you decided to build your custom AI solution? Here\u2019s how we can help\"}),\"\u2026\"]}),/*#__PURE__*/e(\"ul\",{children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",style:{\"--framer-font-size\":\"11px\",\"--framer-text-color\":\"rgb(0, 0, 0)\",\"--framer-text-decoration\":\"none\"},children:/*#__PURE__*/t(\"p\",{children:[\"We\u2019re Fast. By \",/*#__PURE__*/e(i,{href:\"https://artium.ai/insights/ai-and-the-modern-developer\",nodeId:\"a3ufc1L7f\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"incorporating AI into our SDLC\"})}),\", we can deliver your solution much faster.\\xa0\"]})})}),/*#__PURE__*/e(\"ul\",{children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",style:{\"--framer-font-size\":\"11px\",\"--framer-text-color\":\"rgb(0, 0, 0)\",\"--framer-text-decoration\":\"none\"},children:/*#__PURE__*/t(\"p\",{children:[\"We Deliver Quality. By practising TDD and applying our homegrown \",/*#__PURE__*/e(i,{href:\"https://artium.ai/insights/taming-the-unpredictable-how-continuous-alignment-testing-keeps-llms-in-check\",nodeId:\"a3ufc1L7f\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Continuous Alignment Testing\"})}),\" framework, we mitigate the risk of hallucinations without killing the human and creative nature of the AI.\\xa0\"]})})}),/*#__PURE__*/e(\"ul\",{children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"p\",style:{\"--framer-font-size\":\"11px\",\"--framer-text-color\":\"rgb(0, 0, 0)\",\"--framer-text-decoration\":\"none\"},children:/*#__PURE__*/e(\"p\",{children:\"We Enable Adaptability: AI is evolving quickly - by engineering your solution appropriately, we can enable seamless transitions to new models as technologies evolve or need change. Our modular architecture allows for easy integration of new AI models and capabilities, ensuring your investment remains future-proof.\"})})}),/*#__PURE__*/e(\"p\",{children:\"Whether you build or buy, AI is no longer optional to stay competitive in today's market. However, custom AI solutions offer unmatched potential for organisations seeking true differentiation and long-term value. With the right partner and modern development practices, building your own AI-powered software isn't just feasible - it's a fast path to industry leadership.\"})]});export const richText1=/*#__PURE__*/t(o.Fragment,{children:[/*#__PURE__*/e(\"p\",{children:\"When it comes to building AI-powered software, understanding cost is essential for long-term success. In this post, Mike McCormick, VP of Technology at Artium, shares his experience with AI implementations, the common mistakes that companies make, and strategies for managing costs while maintaining high performance.\"}),/*#__PURE__*/e(\"h6\",{children:\"What are the biggest cost drivers in building AI-powered software?\"}),/*#__PURE__*/e(\"p\",{children:\"A lot of the cost drivers in AI-powered software are similar to what you\u2019d expect from building cloud-based applications. You're dealing with the cost of the team doing the build, the cost of infrastructure, and the cost of any COTS products you\u2019ll need along the way. Sticking with the adage \u201Csoftware is to done as grass is to cut,\u201D I prefer to consider these costs as continuous. In addition, when it comes to AI-powered software there are a few unique variables to consider. \"}),/*#__PURE__*/e(\"p\",{children:\"For example, the cost of using hosted foundation models is something that can significantly impact expenses. Right now we are seeing that these models are commoditizing, and vendors are trying to push prices down to stay competitive, but that doesn\u2019t necessarily mean they are affordable within your architecture. If you're making several API calls to an LLM for every user action, those costs can escalate fast. Even if prices drop, high-volume applications can see their expenses soar if they're making multiple calls per interaction. In general, my back-of-the-napkin-math compares API pricing, call volume, and critically of the feature to product improvements against what it costs to have our team create and maintain a more focused model/architecture for ourselves. It often doesn\u2019t take too far beyond a prototype for the focused model/architecture to cost less.\"}),/*#__PURE__*/e(\"p\",{children:\"When we talk about the software tools, Artium has long favored working software in production as a means of driving alignment and product development. In the generative AI era, we find that quick prototypes make a big difference in aligning stakeholders\u2019 understanding and building momentum. It\u2019s easy to prototype something using a hosted model, but when you scale that prototype to users, the cost can scale out of control. In that sense it is important to plan for when you cross over from the convenience of a hosted model to the lower runtime cost of building and hosting your own solutions. \"}),/*#__PURE__*/e(\"p\",{children:\"Hiring AI talent is a significant cost driver. There\u2019s a limited talent pool that can develop, fine-tune, and maintain AI architectures, so whether you're hiring, upskilling internally, or working with consultancies like Artium, the talent that helps you compete and differentiate remains in high demand. These costs appear upfront as salary and transition to retention costs as your talent continuously evaluates growth opportunity, alignment, and the roadmap of cool things you choose to build together. In a highly competitive market engineering leaders have the most leverage on cost by creating environments that retain high-demand individuals. \"}),/*#__PURE__*/e(\"h6\",{children:\"What strategies do you recommend for achieving a balance between high performance and cost efficiency as usage grows?\"}),/*#__PURE__*/e(\"p\",{children:\"The key to balancing performance and cost efficiency is to identify how your AI component delivers value to the user. Too often, companies jump to using LLMs or other AI technologies simply because they\u2019re available, without fully understanding how the technology creates differentiating outcomes or how you intend to track success. When your product team can make statements like \u201Chelping users do this specific task x amount faster converts to y amount of cost savings/revenue/brand reputation/etc.\u201D then it becomes easier for your development team to work out some quick math on how they might build the solution, how that might perform, and what that might cost. Having both the outcome and a napkin scrolled map to achieve the outcome (we have a focused prototyping process to answer these questions) allows you to make decisions on when it is appropriate to apply AI and when you should skip it.\"}),/*#__PURE__*/e(\"h6\",{children:\"What are common mistakes companies make that drive up AI operational costs?\"}),/*#__PURE__*/e(\"p\",{children:\"One of the most common mistakes is underestimating the volume of interactions LLM architectures favor compared to cloud architectures and how that volume relates to long-term cost of hosted foundation models. A good rule of thumb is to consider the LLM a user. When you start asking yourself \u201Chow would a human get to this answer\u201D using natural language and an interrogative process, it becomes clearer that what you think of as a service-to-service call in a cloud architecture might turn into multiple calls to an LLM, which could significantly increase costs if you\u2019re not careful. These hosted foundation models are incredible but they definitely do not come with billing guard rails.\"}),/*#__PURE__*/e(\"p\",{children:\"Another mistake is not preparing for regulatory shifts. As AI continues to receive regulatory attention, companies that don\u2019t plan for compliance can find themselves with unexpected operational costs. I think we\u2019re just beginning to understand how regulations will shape the cost landscape, and I expect that companies in regulated industries will enjoy a brief advantage over traditionally unregulated industries as the latter adjusts to costs, process, and people necessary to remain aligned to regulation.\"}),/*#__PURE__*/e(\"h6\",{children:\"What role does data quality play in the cost of AI maintenance?\"}),/*#__PURE__*/e(\"p\",{children:\"Data quality is a huge factor. For more traditional AI and machine learning models, many data science teams are already familiar with operational processes like creating/maintaining data pipelines and monitoring model drift. But with LLMs, things are still evolving. You can\u2019t just dump data into a model and expect optimal results. You\u2019ll still need to invest in engineering to structure and manage that data properly, which will evolve as your application scales.\"}),/*#__PURE__*/e(\"p\",{children:'At Artium we have a process of Continuous Alignment Testing that we use to make sure new software, new model versions, and new data still play nicely together. We use these tools to boost our confidence and consistency when working with generative models. It costs more to introduce these tools in our continuous integration pipeline but as we know from 30 years of web development, it is better to learn early than to be surprised in production. AI models are not a \"set it and forget it\" technology. You need continuous investment in quality to maintain long-term value.'}),/*#__PURE__*/e(\"h6\",{children:\"How does the choice of AI framework or platform impact long-term costs?\"}),/*#__PURE__*/e(\"p\",{children:\"Your choice of AI framework or platform can have a huge impact on your long-term costs, especially in terms of how locked in you get with a particular vendor. Take OpenAI, for example. Their API makes it really easy to get started, and they\u2019ve done a great job of building in features like function calling or structured data formatting. But all those extra features come with a level of platform dependence.\"}),/*#__PURE__*/e(\"p\",{children:\"If you build heavily on one platform\u2014say, you optimize your app for OpenAI\u2019s GPT models\u2014it\u2019s not a simple lift-and-shift operation to move to another provider like Google\u2019s Vertex or a self-hosted LLM. These models behave differently, and features you rely on in one might not be available in another. That\u2019s why companies need to think about flexibility from day one.\"}),/*#__PURE__*/e(\"p\",{children:\"If you're planning for the long term, you should consider what your migration path might look like. Are you using a third-party model to prove value and get to market quickly, but planning to eventually bring that capability in-house? If so, make sure you\u2019re not building too much lock-in with any one feature set. The more flexible you can make your architecture\u2014whether that\u2019s through building modular systems or leaving room for different models\u2014the more you can control your long-term costs.\"}),/*#__PURE__*/e(\"p\",{children:\"The worst-case scenario is getting locked into a single platform and then discovering that the cost of using that model at scale is unsustainable. Flexibility is key, especially in an iterative field like AI, where the technology and the platforms are evolving rapidly.\"})]});export const richText2=/*#__PURE__*/t(o.Fragment,{children:[/*#__PURE__*/e(\"p\",{children:\"There is a lot of hype around AI and what it can do for businesses. But it\u2019s important that we are realistic about what AI\u2019s current capabilities are and where it can add the most value. In this post, Artium VP of Engineering John Wilger discusses the current state of AI, its potential in the future, and how businesses can use it effectively.\\xa0\"}),/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"How would you describe the current state of AI? What is AI capable of doing today?\\xa0\"})}),/*#__PURE__*/e(\"p\",{children:\"The general public needs to be more aware of what AI can do. Some people fear that AI might take over the world or that AI can solve any conceivable problem. These perspectives are based on misconceptions. We are nowhere close to the idea of a \u201Cgeneralized\u201D artificial intelligence like the ones you see in science fiction, where machines have human-like consciousness and decision-making abilities.\\xa0\"}),/*#__PURE__*/e(\"p\",{children:\"Generative AI is a tremendous statistical engine. When you input a prompt into a system like ChatGPT, it looks at your input, compares it with a massive dataset, and then makes predictions based on patterns in that data. It feels like you are talking to another human a lot of times. But it's often wrong and can give you some persuasive-sounding incorrect information.\"}),/*#__PURE__*/t(\"p\",{children:[\"What excites me about AI right now is how it enhances human-computer interaction. I wrote \",/*#__PURE__*/e(i,{href:\"https://artium.ai/insights/how-generative-ai-is-transforming-user-experience\",nodeId:\"a3ufc1L7f\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"an article\"})}),\" about this in Artium\u2019s blog a while ago. An example is how UX patterns are evolving because we can now interact with systems in ways that don\u2019t just involve filling out form fields and clicking buttons. We\u2019re able to teach an LLM about the functions that our software program can perform rather than having to rely on the human user to decide things like \u201CWhen do I click this button?\u201D or \u201CWhen do I type into this field?\u201D, the LLM is much better at interpreting the interaction on your behalf.\\xa0\"]}),/*#__PURE__*/e(\"p\",{children:\"I do think we'll find more and more uses in the long term, but that\u2019s one area where AI can make a tangible impact today.\"}),/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"Could you make a few predictions on where you think AI is going to go in the future?\\xa0\"})}),/*#__PURE__*/e(\"p\",{children:\"In the near term, AI will primarily serve as an assistant\u2014one that helps us with tasks we're already familiar with, but in a more efficient way. AI can eliminate much of the manual effort in processing data, connecting disparate systems, and making decisions. However, it will still require human oversight to ensure accuracy and relevance. For instance, businesses might use AI to sift through mountains of data and generate reports, but a human will still need to verify those reports and make critical decisions based on them.\"}),/*#__PURE__*/e(\"p\",{children:\"But if we\u2019re looking at the more distant future, that's where things get pretty exciting. Imagine a world where AI isn't just reactive\u2014waiting for us to ask something\u2014but proactive, constantly processing and analyzing incoming data in the background. I envision a future where AI systems can generate insights autonomously without needing a prompt from a human. For example, an AI could monitor real-time sales data and inventory in an e-commerce business and recommend to a business leader, saying, \\\"Based on the trends I'm observing, you should consider marketing Product X this week.\\\" That level of proactive decision-making is where I think AI will make a massive leap in the future.\"}),/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"How do you think AI is going to fall short?\\xa0\"})}),/*#__PURE__*/e(\"p\",{children:\"AI\u2019s current limitations remind me of the early days of the internet, especially during the dot-com bubble. Back then, there was an incredible amount of hype around the web, but eventually, that bubble burst. What remained were the foundational pieces that created real value, like the e-commerce platforms we take for granted today.\"}),/*#__PURE__*/e(\"p\",{children:\"I think AI will experience something similar. We\u2019re in the middle of an AI bubble, and many people are rushing to use this new technology without fully understanding it. At some point, there will be a \u201Ccorrection,\u201D where the excitement will settle, and we\u2019ll be left with truly valuable applications of AI\u2014those that make our lives easier and our work more efficient.\"}),/*#__PURE__*/e(\"p\",{children:\"In the short term, AI will fall short of this Star Trek science fiction dream many have, where it solves all our problems autonomously. But I don't see this as a permanent failure; it's just part of the growth process.\"}),/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"With the rapid advancements in AI, how do you assess the feasibility of solving certain business problems using AI, and how do you manage client expectations about AI\u2019s limitations?\"})}),/*#__PURE__*/e(\"p\",{children:\"When a client comes to us with a business problem, I first assess whether AI can realistically solve it. The critical question is: \u201CIs this a problem that can be best solved by analyzing data and making connections between facts?\u201D If there\u2019s a pattern in the data, then AI can help. AI is an excellent fit if the problem involves repetitive decision-making or something where historical data can guide predictions\u2014like identifying customer preferences or automating workflows.\"}),/*#__PURE__*/e(\"p\",{children:\"But I also make it clear to clients that AI isn\u2019t going to invent solutions out of thin air. It\u2019s not magic. AI might not be the right tool if the problem requires brand-new thinking or if there\u2019s no clear data pattern to draw from. I spend a lot of time managing client expectations, especially given the hype around AI. I explain that AI can be a potent tool, but it\u2019s still a tool. It can help speed up processes and assist in decision-making, but it won\u2019t replace human intuition or creativity. Setting these boundaries helps clients understand what AI can achieve for their business.\"}),/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"Data quality is often cited as a major cause of AI project failure. How do you ensure that your clients have the right data infrastructure and governance in place to support successful AI projects?\"})}),/*#__PURE__*/e(\"p\",{children:\"Data is fundamental to AI\u2019s success, and many AI projects fail because of poor data quality or infrastructure. Historically, the industry taught software developers to minimize data retention because storage was expensive. We trained developers to build systems that overwrite or delete old data simply because there wasn\u2019t enough room to store everything. But today, storage is cheap, and data has become one of the most valuable assets for any business.\"}),/*#__PURE__*/e(\"p\",{children:\"The critical shift that\u2019s happening\u2014and what I emphasize to clients\u2014is the move toward event-sourced systems. Instead of just tracking the current data state, we need to capture the entire sequence of events that lead to a particular outcome. For instance, in an e-commerce setting, don\u2019t just record that a customer bought a product\u2014record every interaction that led up to that purchase. Did they add the item to their cart and then remove it? Did they view the product multiple times before buying? All of these events provide crucial context for understanding customer behavior.\"}),/*#__PURE__*/e(\"p\",{children:\"Event-driven data allows AI systems to analyze not only what happened but why it happened. And that\u2019s where AI becomes incredibly powerful. It can connect all those dots and offer far more accurate and meaningful insights. The most important takeaway here is: Don\u2019t lose data. Capture as much as you can because you never know what will be helpful to AI in the future. Storage is cheap, but the value of the data is priceless.\"})]});export const richText3=/*#__PURE__*/t(o.Fragment,{children:[/*#__PURE__*/t(\"p\",{children:[\"AI is set to revolutionize many aspects of software products and development. However, one thing that has not changed is the fundamentals of \",/*#__PURE__*/e(\"strong\",{children:\"Product-Market Fit\"}),\". You still need to identify an opportunity that genuinely matters to users, develop a strategic product roadmap, and iteratively deliver something they can't live without.\"]}),/*#__PURE__*/e(\"p\",{children:\"Amid the current AI hype, it's common to see companies rush into developing innovative AI products without fully understanding what AI is capable of doing or taking the essential steps to establish customer alignment.\"}),/*#__PURE__*/e(\"p\",{children:\"In this post, Mark Whaley, Head of Product at Artium, shares his insights on building AI products that are not just viable but truly valuable.\"}),/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"What are the biggest challenges in defining a viable AI product roadmap?\"})}),/*#__PURE__*/e(\"p\",{children:\"With any roadmap (not just AI), the biggest challenge I see is the misalignment between business goals and how to implement them\u2014clarifying the solution you want to build and figuring out the best way to build it.\"}),/*#__PURE__*/e(\"p\",{children:\"One common pitfall, especially with AI, is the gap between what people read in the news and what's actually feasible and helpful for a business. Business owners sometimes get fixated on the latest tech trends, like large language models (LLMs), when they could be solving that same problem rather simply with a traditional tech build. If you're paying attention to what the user really needs then oftentimes you may not even need AI. \"}),/*#__PURE__*/e(\"p\",{children:\"The other piece that I want to highlight is that even if AI might be a good solution for what you want to build, you may not have the knowledge or resources to build it properly. For example, there are a number of different AI architectures that can be implemented for different use cases. Do you want a simple AI chat, are you transforming data with AI assistance, or are you going for a more complex multi-model generative interface? Many businesses don\u2019t realize the magnitude of these fundamental decisions. \"}),/*#__PURE__*/e(\"p\",{children:\"In short, to get a clear roadmap you have to take into account the capabilities of AI combined with tangible business goals, the infrastructure and data in place to back it up.\"}),/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"How do you balance innovation with practicality when selecting AI projects to pursue?\"})}),/*#__PURE__*/e(\"p\",{children:\"I will always prioritize practical applications with a clear path towards deployment and a measurable business impact. I think that a lot of times leaders find themselves hooked by the allure of a novel tech advancement, but they may run into practical constraints like time, money, and customer needs. Innovation is exciting, but if you chase it for its own sake, you can end up with solutions that are too complex or don\u2019t really solve the customer\u2019s problem.\"}),/*#__PURE__*/e(\"p\",{children:\"My advice is to always start with an AI project that addresses well-defined customer pain points and delivers measurable outcomes. This will allow you to set realistic timelines and investment thresholds. For example, choosing to build an AI solution for optimizing appointment scheduling might be less glamorous than something like a cutting-edge disease diagnosis system, but it\u2019s much more practical and can deliver immediate business value.\"}),/*#__PURE__*/e(\"p\",{children:\"When coming up with a new implementation for AI, I\u2019d suggest you start small, solve a tangible problem, then build from there. Once you have success with a smaller, practical solution, you can expand into more ambitious projects.\"}),/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"What strategies do you use to validate that an AI solution is truly adding value to the customer?\"})}),/*#__PURE__*/e(\"p\",{children:\"Again, a good product manager focuses on solving specific, well-defined issues. This means making sure the AI solution is tied directly to real-world customer needs.\"}),/*#__PURE__*/e(\"p\",{children:\"You validate by collecting both quantitative and qualitative feedback as you build. You want to measure AI's impact on efficiency, user experience, and revenue impact. Data drives your decisions, and that data should come from both customer interviews and how people are actually interacting with the product.\"}),/*#__PURE__*/e(\"p\",{children:\"For example, if you make a change in a prompt that you're writing for a LLM, you would want to immediately measure what that's doing to your customer experience and its impact on how they're moving through your product. Use frequent iterations where the models are tested then refined based on that measurement.\"}),/*#__PURE__*/e(\"p\",{children:\"Take a chatbot as an example. It may be technically impressive but it may still receive poor feedback from users. Chatbots can be challenging to talk to, hard to navigate or not really helpful for specific queries. Users end up getting fed up with them and just go to a website FAQ or ask to talk to a real person. To fix this, companies can refine their chatbot experience by adding in natural language processing capabilities and train the model using real customer data. Through that process the experience will get better over time.\"}),/*#__PURE__*/e(\"p\",{children:\"Adopting a \u201Ccustomer mindset\u201D is something everyone talks about, but hardly anyone really does. I\u2019ve seen this in larger organizations that claim to be customer-centric but don\u2019t actually conduct interviews or measure the impact of their decisions on real users. AI-driven products need continuous refinement, so it\u2019s critical to make user feedback and continuous discovery a part of your development lifecycle. \"}),/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"How do you integrate user feedback into the development process for AI-driven products? What key metrics do you use to measure the success of AI initiatives?\"})}),/*#__PURE__*/e(\"p\",{children:\"The key is to involve users early and often. Many teams make the mistake of building first and asking for feedback later, but we prefer to integrate user feedback right from the ideation phase. That way, the product evolves with the customer in mind from the very beginning.\"}),/*#__PURE__*/e(\"p\",{children:\"We use beta testing, pilot programs, early prompt testing, and iterative development cycles to refine our solutions based on real-world feedback loops. Metrics like customer satisfaction, AI model accuracy, and business performance are vital. But some metrics are more intuitive\u2014like whether a chatbot feels human, which can drive engagement.\"}),/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"What role does cross-functional collaboration play in the success of AI projects, and how do you foster this approach at Artium?\"})}),/*#__PURE__*/e(\"p\",{children:\"Cross-functional collaboration is the cornerstone of what we do at Artium. AI projects are inherently multidisciplinary, requiring input from engineers, AI architects, product managers, designers, and business stakeholders. You need this blend of perspectives to ensure a holistic approach to building AI solutions.\"}),/*#__PURE__*/e(\"p\",{children:\"At Artium, we emphasize the importance of frequent, clear communication across teams. We don\u2019t throw a roadmap over the fence for the engineering team to execute; instead, we work together continuously. Regular touchpoints, shared artifacts, and a culture of transparency ensure everyone stays aligned.\"}),/*#__PURE__*/e(\"p\",{children:\"For example, if we\u2019re developing an AI-powered recommendation feature for a mobile app, it\u2019s not just the engineers focusing on the technical aspects. Designers ensure the user experience is seamless, AI architects ensure reliability, alignment, and accuracy, and the product managers ensure business goals are met. If any one of these elements are out of sync, the project will fail to meet user expectations.\"}),/*#__PURE__*/e(\"p\",{children:\"We also hold weekly retrospectives to adapt based on feedback and ensure we\u2019re always focused on delivering rapid tangible outcomes.\"})]});export const richText4=/*#__PURE__*/e(o.Fragment,{children:/*#__PURE__*/e(\"h6\",{children:\"Interested to learn more about best practices in defining a viable and valuable AI product?\"})});export const richText5=/*#__PURE__*/t(o.Fragment,{children:[/*#__PURE__*/e(\"p\",{children:\"Many enterprise companies are jumping to create new products and features centered around the tech world\u2019s hottest topic: Artificial Intelligence. \"}),/*#__PURE__*/e(\"p\",{children:\"However, 80% of these initiatives reportedly fail. Many assume that a few lines of code or a quick integration of an AI model will get them to where they need to be, but the truth is far more challenging. \"}),/*#__PURE__*/e(\"p\",{children:\"In this post, Artium Senior Software Engineer David Wiszowaty will dive into what sets truly successful AI products apart from those that fail - and share insights on how to do it. \"}),/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"Q: What are the most common misconceptions about AI in products that you've encountered?\"})}),/*#__PURE__*/e(\"p\",{children:\"One of the biggest misconceptions I've encountered while building AI products, especially on the engineering side, is the oversimplification of what it takes to actually build AI features. Many people think that developing an AI application is as straightforward as integrating an LLM, writing a few prompts, and voil\\xe0\u2014you have a fully functional AI product. \"}),/*#__PURE__*/e(\"p\",{children:\"While this approach can get you partially the way there, AI development is much more complex. Creating a truly valuable and reliable AI product means diving deep into fine-tuning the model, making incremental changes, and sometimes even incorporating multiple AI agents that work together to deliver the desired outcomes. There's also the need for constant evaluation and improvement. AI doesn't always deliver reliable or accurate results right out of the box, and it requires a lot of trial and error to get it to a state where it can consistently provide value to end users.\"}),/*#__PURE__*/e(\"p\",{children:\"For instance, when you're developing an AI model, there are many different aspects you need to consider: data preprocessing, model training, hyperparameter tuning, integration with existing systems, and continuous monitoring.\"}),/*#__PURE__*/e(\"p\",{children:\"This misconception of AI as an easy solution can lead to unrealistic expectations and disappointment when the AI doesn\u2019t immediately perform as hoped. It\u2019s crucial to recognize that AI development is an ongoing process that involves learning, adjustment, and refinement.\"}),/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"Q: How does Artium\u2019s practices play a role in building a successful AI product? \"})}),/*#__PURE__*/e(\"p\",{children:'There are several pieces to this, but one of the core practices we focus on is testing\u2014specifically Test-Driven Development (TDD). Since AI is inherently non-deterministic, meaning that the responses you get can be random and unpredictable. This can make testing seem tricky at first. In my current project we are relying heavily on testing how our AI functionality behaves. The way that we are approaching this is through a practice called \"Continuous Alignment Testing\" (CAT). This testing framework allows us to validate the AI behavior by providing it with numerous scenarios and assessing the AI responses for tone, accuracy, and efficacy.  In the case of my current project, the AI model searches for people with specific skill sets and summarizes the results. We\\'ve written tests to ensure that when given a specific prompt, the AI returns reliable results and doesn\u2019t hallucinate or make up information. '}),/*#__PURE__*/e(\"p\",{children:\"Another practice that we do at Artium that has been super valuable for developing successful AI features is Pair Programming. Since the responses we get from AI models are non-deterministic, there are numerous different situations we can be in. Being able to pair allows us to bounce ideas off each other in real time. We can then try out these scenarios and figure out together how to update the AI prompt to better handle every scenario. Pairing and even mobbing can increase and expand the number of different approaches to try and improve the AI feature.\"}),/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"Q: What key factors differentiate successful AI products from those that fail?\"})}),/*#__PURE__*/e(\"p\",{children:\"This is a great question and one that took me some time to think about. I believe one of the key differentiators is how you approach building AI components or features. A common pitfall is treating AI simply as a chatbot or an assistant that only responds to commands. While there is value in chatbots, it isn\u2019t the most innovative AI implementation.\"}),/*#__PURE__*/e(\"p\",{children:'On the project I am currently working on, we\u2019re approaching AI as more of a \"teammate\" than just an assistant. A teammate works alongside you to achieve a shared goal, not just reacting to commands, but proactively participating in the workflow. A great example of this is GitHub Copilot, which positions itself as a coding partner and \u201Cpairing buddy\u201D rather than a simple assistant. Using Github Copilot, a developer will go about their regular, day to day workflow, and occasionally the Copilot will contribute and make suggestions where it thinks it makes sense. It\u2019s not always perfect, but there are plenty of occasions where the Copilot provides valuable results that the developer wouldn\u2019t have thought of entering into a chatbot interface. This kind of approach can lead to much more valuable and impactful AI implementations.'}),/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"Q: What advice would you give to other engineers looking to improve the success rate of their AI projects?\"})}),/*#__PURE__*/e(\"p\",{children:\"The biggest piece of advice I\u2019d offer is to think about testing very early on in the development process. It is very time-consuming and difficult to gauge how successful a prompt is in assuring it is responding appropriately. This becomes more apparent when making any sort of modification to the AI prompt. Any change to the prompt can cause the AI to behave very differently in some scenarios. Without automated tests and monitoring, it becomes very difficult to know if the prompt change actually broke any existing behavior. \"}),/*#__PURE__*/e(\"p\",{children:\"Another piece to consider is to think about improving visibility into what is happening behind-the-scenes with these AI components. When developing complex AI features, it can all feel very magical and tough to understand. You tend to ask yourself, \u201CIs it actually working as intended?\u201D Logging and understanding what prompts are being sent and how the AI is processing them can help you identify where things might be going wrong and where adjustments are needed.\"}),/*#__PURE__*/t(\"p\",{children:[\"Finally, choose the right tools and libraries that offer good visibility into what\u2019s happening under the hood. Avoid libraries that abstract away too much of the inner workings, as they make it harder to understand what\u2019s happening and how to debug issues when they arise. Opt for lighter wrappers or open-source solutions where you can see and understand the interactions between components. In my current project, it was very helpful to experiment with different libraries since each of these libraries offer different pieces of AI functionality. We were able to notice that some AI libraries do better in some specific workflows, while not so great in other workflows. There were numerous libraries and frameworks that we experimented with, such as \",/*#__PURE__*/e(i,{href:\"https://www.langchain.com/\",nodeId:\"a3ufc1L7f\",openInNewTab:!0,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"LangChain\"})}),\", \",/*#__PURE__*/e(i,{href:\"https://www.crewai.com/\",nodeId:\"a3ufc1L7f\",openInNewTab:!0,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"CrewAI\"})}),\", \",/*#__PURE__*/e(i,{href:\"https://www.llamaindex.ai/\",nodeId:\"a3ufc1L7f\",openInNewTab:!0,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"LlamaIndex\"})}),\", and just using the \",/*#__PURE__*/e(i,{href:\"https://cloud.google.com/vertex-ai/docs/python-sdk/use-vertex-ai-sdk\",nodeId:\"a3ufc1L7f\",openInNewTab:!0,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Vertex AI SDK\"})}),\". Ultimately, we chose to use LlamaIndex because it made it easy to implement RAG (Retrieval-Augmented Generation) and build a multi-agent AI architecture. Additionally, it provided the right balance of abstraction and transparency, making it easier to understand what happens behind the scenes.\"]})]});export const richText6=/*#__PURE__*/e(o.Fragment,{children:/*#__PURE__*/e(\"h6\",{children:\"Interested in learning more about building successful AI features?\"})});export const richText7=/*#__PURE__*/t(o.Fragment,{children:[/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"Q: How do you see the future of AI user experiences evolving beyond the current chat interface paradigm? What innovative applications of AI do you believe could redefine how users interact with technology?\"})}),/*#__PURE__*/e(\"p\",{children:\"It\u2019s true that right now many people see AI experiences primarily through chatbots, but there are a lot of exciting developments beyond that. From my perspective as a developer, I expect we will see a continued improvement of the copilot experience. We\u2019ll have super intellisense that will autocomplete and suggest code as we type. Although it\u2019s still evolving and sometimes makes incorrect suggestions, its potential to improve as it gains more context about your entire project is significant.\"}),/*#__PURE__*/e(\"p\",{children:\"At Artium, we believe in the value of pair programming, where two developers work together on the same codebase. An AI partner could be invaluable when no one else is available to pair with you. Imagine talking to an AI as you would with a human partner over Zoom, where it could act as either the Navigator or the Driver in the coding process.\\xa0\"}),/*#__PURE__*/e(\"p\",{children:\"As the Navigator it will speak to you and analyze your code as you write it to better understand what you're building. It\u2019ll stop you when you start to go down a rabbit hole or if you make a mistake while coding. It\u2019ll keep you on track.\\xa0\"}),/*#__PURE__*/e(\"p\",{children:\"As a Driver it will be in control of the code that you are writing. We will navigate it by talking to it and describing what step we think we should take next.\\xa0\"}),/*#__PURE__*/e(\"p\",{children:\"Ideally this experience would be latency free and as accurate (or more so) than a real life pair programming session.\"}),/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"Q: What are the most significant technical hurdles in developing AI systems that can understand and reason with human language? How can engineers overcome these challenges to create seamless and intuitive user experiences?\"})}),/*#__PURE__*/e(\"p\",{children:\"If you asked me this a couple years ago I would say the machine learning part is the biggest hurdle. That\u2019s all changed now with LLMs and all the tools available for both the developer trying to build the thing and on the user side.\\xa0\"}),/*#__PURE__*/e(\"p\",{children:\"As for building AI systems into products for users, the biggest technical hurdle that I see is with the reliability of responses. We know LLMs are inherently non-deterministic, but that doesn\u2019t mean we can\u2019t get useful responses, it just means it takes some work. Whether it\u2019s a chat interface, an AI programming assistant, or a backend agent, ensuring accurate and reliable outputs is crucial. At Artium, we use something called Continuous Alignment Testing (CAT). This involves automated testing that runs in our CI pipeline, checking the accuracy of AI responses to specific prompts. For example, if the AI is asked to list products related to entertainment, it shouldn\u2019t return irrelevant items like cookware. These tests help us maintain and improve the quality of AI outputs by providing feedback on how changes in the code affect accuracy. You don\u2019t want a product you build to start hurling insults or misinformation at customers. You need to be able to trust the kinds of responses tools we build will give.\"}),/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"Q: How do you ensure that the AI-driven systems you build are truly user-centric, and what are the key principles that guide this process?\"})}),/*#__PURE__*/e(\"p\",{children:\"This is where collaboration with our product management team is crucial. It starts with understanding user needs through interviews and experiments, which are then translated into requirements. These requirements flow through the entire system, from ideation to implementation, and all the way to deployment and user feedback. It\u2019s a continuous process of refining and adapting based on both technical challenges and evolving user needs. A strong product team in collaboration with the development team is essential in ensuring that what we build resonates with users and meets their expectations.\"}),/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"Q: As AI continues to advance, how can engineers balance the desire for cutting-edge user experiences with the need for system reliability and user trust? What strategies do you recommend for maintaining this balance?\"})}),/*#__PURE__*/e(\"p\",{children:\"It\u2019s all about trusting the fundamentals of good software engineering. While AI brings new possibilities, the old-school principles of software development still apply. We still practice pair programming because two heads really are better than one when it comes to writing code. Having a tight relationship between the Product, Design, and Engineering teams keeps churn lower and helps keep code quality high. These teams are agile, able to experiment, fail, and quickly fix issues. Incorporating quality and safety measures into this process ensures that we can deliver reliable software without unnecessary delays. As much as AI is changing things, these fundamentals are what\u2019s going to let teams build new experiences users can trust and rely on.\"})]});export const richText8=/*#__PURE__*/e(o.Fragment,{children:/*#__PURE__*/e(\"h6\",{children:\"Interested in learning more about building apps that incorporate AI-driver user experiences?\"})});export const richText9=/*#__PURE__*/t(o.Fragment,{children:[/*#__PURE__*/t(\"p\",{children:[\"Artificial intelligence is poised to revolutionize user experiences across many industries. Two industries in particular that we\u2019ve seen to be early adopters in AI are Healthcare and Media & Entertainment. \",/*#__PURE__*/e(i,{href:\"https://www.linkedin.com/in/charlotteonsrud/\",nodeId:\"a3ufc1L7f\",openInNewTab:!0,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Charlotte Onsrud\"})}),\", Director of Strategy & Innovation in Healthcare and M&E at Artium shares her thoughts on how AI is transforming user experiences for the companies she\u2019s been working with and what the future holds for businesses across industries ready to embrace AI.\"]}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"strong\",{children:\"Q: In what ways can AI technology revolutionize user experiences?\"})}),/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(\"strong\",{children:\"A:\"}),\" AI is already making user experiences so much more dynamic. Instead of interacting with an application through traditional input fields, AI enables users to engage in a way that feels more like a conversation with a person. You can use natural language to ask questions or request actions, making the experience feel more intuitive and human. Rather than toggling different fields or pressing buttons, you're having a more fluid, interactive experience. That, to me, is the biggest impact GenAI will have\u2014it brings a more human component to interacting with software.\"]}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"strong\",{children:\"Q: What are some use cases you're hearing from leaders in the Media and Entertainment (M&E) and Healthcare?\\xa0\"})}),/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(\"strong\",{children:\"A: \"}),\"The primary focus is on two things: search and insights.\\xa0\"]}),/*#__PURE__*/e(\"p\",{children:\"In terms of search, AI is making it easier and more enjoyable for consumers to find the right content or what they want to purchase. For employees, AI is helping them find the resources they need to do their jobs more effectively.\\xa0\"}),/*#__PURE__*/e(\"p\",{children:\"For leaders, it's about gaining better insights and predictive analytics into their business operations. These advancements enable quicker, more informed decision-making and a deeper understanding of their industry landscapes.\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"strong\",{children:\"Q: What industries do you think are more likely to disrupt through innovative AI user experiences?\"})}),/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(\"strong\",{children:\"A:\"}),\" I believe every industry has the potential to disrupt through AI. However, the most appetite and opportunity I\u2019ve seen so far has been within Healthcare. Healthcare is incredibly broad, and the applications for AI are vast. It's easy to see the positive impact AI can have in this field. For instance, I was talking to someone recently who mentioned that 70% of a physician's time is spent on paperwork. If AI could automate even half of that admin work, it would drastically increase the amount of time physicians can spend with patients, which could lead to significantly improved patient care.\"]}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"strong\",{children:\"Q: Can you share any success stories of companies that have effectively implemented advanced AI user experiences?\"})}),/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(\"strong\",{children:\"A:\"}),\" With large language models (LLMs) specifically, we\u2019re still in the early stages of being able to determine the impact our solutions will deliver for our clients. A lot of what we're working on is under NDA, so I can't share specifics, but what I can say is that, as we progress through these engagements and demonstrate our progress to clients, they\u2019re getting more and more excited about the potential for increased efficiencies, monetization and other forms of value add to their end customers\\xa0\"]}),/*#__PURE__*/e(\"p\",{children:\"Another important point to make here is how it\u2019s never been this easy to deliver so much, so fast. The speed at which we're able to build these solutions is truly unprecedented. This rapid delivery and the promise of even more advanced capabilities in the near future also has many companies eager to explore what's possible with AI.\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"strong\",{children:\"Q: What do you envision as the next big breakthrough in AI-driven user experiences?\"})}),/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(\"strong\",{children:\"A:\"}),\" I think we\u2019re going to start seeing more multimodal features come into play , where people can interact with applications not just through text, but also\\xa0 voice, image, and video.\\xa0\"]}),/*#__PURE__*/e(\"p\",{children:\"The significance here being that, if implemented correctly, we will see new ways for users to engage with software, better accessibility for all and new monetization opportunities for tech forward companies.\\xa0\\xa0\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"strong\",{children:\"Q: How can businesses stay ahead of the curve in adopting and integrating next-gen AI user experiences?\"})}),/*#__PURE__*/t(\"p\",{children:[/*#__PURE__*/e(\"strong\",{children:\"A:\"}),\" The landscape is evolving so quickly that it\u2019s important for businesses to start experimenting with AI right away. If you're waiting for the hype or pace of change to slow down, you\u2019re going to find yourself out of touch and behind the curve very quickly.\\xa0\"]}),/*#__PURE__*/e(\"p\",{children:\"As we always tell our people and our clients, the best way to learn is through doing so, if you don\u2019t know where to start, pick an area of need or interest and start rapidly iterating. The key is to stay open minded, curious and agile, especially as the tech continues to evolve. It may not always work out as planned, but that\u2019s true for all software development (and life!).\\xa0\"}),/*#__PURE__*/e(\"p\",{children:\"And if you really find yourself stuck then, well, give us a call! We\u2019d love to help.\"})]});export const richText10=/*#__PURE__*/t(o.Fragment,{children:[/*#__PURE__*/e(\"p\",{children:\"Have you ever poured your heart and soul into a new feature, only to discover it didn't quite hit the mark with your users? This is a common challenge in product development, and often, the root cause is a disconnect between what we think users want and what they need. This is where genuinely understanding the voice of the customer becomes essential. \"}),/*#__PURE__*/e(\"p\",{children:\"The voice of the customer is essentially feedback from your users about their experiences and expectations. It's more than just data; it's a comprehensive insight into what users think, feel, and need from your product. When you truly understand what users want, you can prioritize features that add genuine value and avoid investing in those that miss the mark.\"}),/*#__PURE__*/t(\"p\",{children:[\"AI-powered tools like \",/*#__PURE__*/e(i,{href:\"https://openai.com/index/hello-gpt-4o/\",nodeId:\"a3ufc1L7f\",openInNewTab:!0,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"ChatGPT 4o \"})}),\"can help with this by simulating user interactions. For example, natural language processing can mimic conversations with real users, enabling us to explore their needs, preferences, and challenges realistically. \"]}),/*#__PURE__*/e(\"p\",{children:\"What sets ChatGPT apart from other tools is its ability to facilitate multi-modal conversations using voice, creating an experience resembling a real-life phone conversation with a real user. This level of immersion allows us to gather authentic feedback and gain a deeper understanding of user perspectives.\"}),/*#__PURE__*/e(\"p\",{children:\"Nothing will beat true customer interaction. But if you create a series of personas and have a handful of conversations with them, you will make your valuable time interacting with your customers much more effective.\"}),/*#__PURE__*/e(\"h5\",{children:\"Benefits of Using AI for User Feedback\"}),/*#__PURE__*/e(\"p\",{children:\"When it comes to cost-effectiveness, ChatGPT shines. Think about the traditional user feedback methods \u2013 recruitment, scheduling, participant compensation, and all the logistical headaches that come with them. These can quickly add up to a significant expense in money and time. But with ChatGPT, you can bypass many of these costs and challenges. It's a lean, efficient way to gather extensive user insights without needing a large budget or a dedicated team of user researchers. This means you can allocate your resources more strategically, focusing on what matters: developing a product that resonates with your users. It's like having a direct line to your users' thoughts and preferences without the usual barriers and expenses. This cost-effectiveness makes ChatGPT an invaluable tool for startups, small teams, and anyone looking to maximize their user research ROI.\"}),/*#__PURE__*/e(\"p\",{children:\"Scalability is another significant advantage that ChatGPT brings to the table. This tool allows you to run multiple scenarios and interactions simultaneously, dramatically speeding up the feedback and iteration. This means you can handle various user personas and scenarios at a much larger scale, allowing you to gather more insights in less time. This translates to a more efficient and streamlined product development process, as you can quickly incorporate user feedback and make data-driven decisions. It's like having a team of user researchers working around the clock without the associated costs and logistical challenges. This scalability empowers you to keep your finger on the pulse of user needs and preferences, ensuring that your product development remains agile and responsive.\"}),/*#__PURE__*/e(\"p\",{children:\"The authenticity of conversational feedback with ChatGPT, primarily through voice interactions with GPT-4o, enhances the depth of insights collected. Simulating genuine conversations captures the words and emotional nuances behind user responses. This level of detail can reveal subtle preferences and pain points that might be missed in traditional surveys or interviews.\"}),/*#__PURE__*/e(\"p\",{children:\"Furthermore, the iterative nature of integrating ChatGPT into the agile process enables continuous improvement. Updating personas and scenarios based on ongoing feedback keeps the development process aligned with evolving user needs. This iterative feedback loop supports the core principles of agile methodologies, promoting constant enhancement and user-focused development.\"}),/*#__PURE__*/e(\"p\",{children:\"Bringing the voice of the customer into your development process involves several key steps. You can ensure your product meets real user needs by carefully creating user personas, developing realistic scenarios, and utilizing ChatGPT for simulated conversations. Here's how to do it:\"}),/*#__PURE__*/e(\"h5\",{children:\"Step-by-Step Guide to Using ChatGPT for User Feedback\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"strong\",{children:\"Step 1: Define and Create Detailed User Personas\"})}),/*#__PURE__*/e(\"p\",{children:\"The first step is to define and create detailed user personas. A good persona includes demographics, behaviors, goals, and challenges that represent your typical user. ChatGPT can be a valuable tool, helping you generate realistic and comprehensive user personas through interactive prompts.\"}),/*#__PURE__*/e(\"p\",{children:\"To create a user persona with ChatGPT, provide the AI with a brief description of your target user and the product you're developing. \"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"strong\",{children:\"Step 2: Set Up ChatGPT for Simulated Conversations\"})}),/*#__PURE__*/e(\"p\",{children:\"Once the personas and scenarios are ready, set up ChatGPT for simulated conversations. Craft prompts that are specific and realistic to the scenarios.\"}),/*#__PURE__*/e(\"p\",{children:\"This setup ensures the simulated conversation is as close to real user interaction as possible. \"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"strong\",{children:\"Step 3: Conduct Simulated Interviews with GPT-4o Voice\"})}),/*#__PURE__*/e(\"p\",{children:\"Take your simulated interviews to the next level by utilizing the GPT-4o voice feature. This innovative technology allows you to engage in near-real-time voice conversations with your ChatGPT personas, creating an experience that closely mimics a live user interview.\"}),/*#__PURE__*/e(\"p\",{children:\"To get started, load your persona into ChatGPT and switch to voice interaction mode on the phone using the same thread.\"}),/*#__PURE__*/e(\"p\",{children:\"The GPT-4o voice's near-instantaneous response time allows for a smooth, natural conversation flow, closely replicating a real-life interview. You can dive deeper into specific topics, ask follow-up questions, and gather rich, nuanced feedback that text-based interactions might miss.\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"strong\",{children:\"Step 4: Analyze and iterate\"})}),/*#__PURE__*/e(\"p\",{children:\"Continuously synthesize feedback and refine your product based on user insights.\"}),/*#__PURE__*/e(\"p\",{children:/*#__PURE__*/e(\"strong\",{children:\"The video below will show you exactly how to do this:\"})}),/*#__PURE__*/e(n.div,{className:\"framer-text-module\",style:{\"--aspect-ratio\":\"560 / 315\",aspectRatio:\"var(--aspect-ratio)\",height:\"auto\",width:\"100%\"},children:/*#__PURE__*/e(a,{componentIdentifier:\"module:NEd4VmDdsxM3StIUbddO/bZxrMUxBPAhoXlARkK9C/YouTube.js:Youtube\",children:t=>/*#__PURE__*/e(r,{...t,play:\"Off\",shouldMute:!0,thumbnail:\"Medium Quality\",url:\"https://www.youtube.com/watch?v=qYKmKs1hTL4&t=209s\"})})})]});export const richText11=/*#__PURE__*/e(o.Fragment,{children:/*#__PURE__*/e(\"h6\",{children:\"Need help implementing AI into your user feedback process?\"})});export const richText12=/*#__PURE__*/t(o.Fragment,{children:[/*#__PURE__*/e(\"ol\",{children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"h6\",children:/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"Can you describe your background and how the software development landscape has evolved since you got started in this industry?\"})})})}),/*#__PURE__*/e(\"p\",{children:\"I've been in software development professionally for about 25 years. The first program I ever wrote in BASIC when I was 8 years old was to get a computer to have a conversation with me. That's all I ever wanted was to be able to converse with my computer. I've always been very interested in AI, and it's been a driving force throughout my career.\"}),/*#__PURE__*/e(\"p\",{children:\"When I started, the landscape was completely different. Programming was at a very low level and it was very difficult to get anything done. There was no agile methodology, no object-oriented programming, and the internet was barely accessible to the public. \"}),/*#__PURE__*/e(\"p\",{children:\"Over the years, layers of abstraction have made it easier and more intuitive to create, interact, and program. With generative AI, we can finally give high-level commands, and the system figures out the details. This shift from low-level programming to high-level abstraction is the most significant change I've witnessed.\"}),/*#__PURE__*/e(\"ol\",{start:\"2\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"h6\",children:/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"How has the inclusion of AI in products changed the way product teams operate?\"})})})}),/*#__PURE__*/e(\"p\",{children:\"To answer that question is hard because you're asking it as if it's already happened. Right now we're really in the middle of it. But there are some specific changes in progress that I see going on right now.\"}),/*#__PURE__*/e(\"p\",{children:\"Creating products that include AI in the center means taking a different approach altogether. For example, traditionally when building software we build the smallest working version of the product first and then we iterate on it, adding features and capabilities. Somewhere down the line we create dashboards that administrators can use to manage the system. AI-based software flips this on its head.\\xa0\"}),/*#__PURE__*/e(\"p\",{children:\"We have to create dashboards right up front that allow us to give rapid feedback on how the LLM or diffusion model is interpreting what we ask of it. As an LLM can take any input in multiple modalities, that means a lot of variation from one project to the next. So a software team has to build testing frameworks and dashboards first.\\xa0\"}),/*#__PURE__*/e(\"p\",{children:\"Another significant change is that balanced teams used to consist of design, product and engineering. AI-based applications have added a fourth leg to that stool: AI. AI affects how the business approaches the problem, how the UI is designed and built and the architecture of the app. Not having a person in the role to advocate for AI, the way product advocates for business, designers represent users and engineers champion systems, is a sure path to a half-baked AI effort.\\xa0\"}),/*#__PURE__*/e(\"p\",{children:\"There are also a lot of examples of how AI is changing how product teams operate whether they are building AI-products or something more traditional. Lots of other examples exist such as, AI helping product teams by handling time-consuming tasks like data analysis and summarization, allowing product managers to focus more on strategy and user engagement. So what happens here is the AI helps by augmenting the human intelligence, and then the humans use that and augment the AI. They're both collaborating to get to a shared goal and a shared vision.\\xa0\"}),/*#__PURE__*/t(\"p\",{children:[\"The AI tool that we built at Artium called \",/*#__PURE__*/e(i,{href:\"https://artium.ai/apex\",nodeId:\"a3ufc1L7f\",openInNewTab:!0,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"APEX\"})}),\" can actually take an initial idea for a new product, expand it into a comprehensive plan, and then distill it back into actionable user stories for engineers to build. This process has streamlined our workflow and made product management way more efficient. \"]}),/*#__PURE__*/e(\"ol\",{start:\"3\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"h6\",children:/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"What tools do you believe are crucial for product managers to be successful in working with AI-driven products?\"})})})}),/*#__PURE__*/t(\"p\",{children:[\"It depends on if the PMs are technical or not. For technical PMs, getting familiar with data handling tools can be very useful in better understanding the users needs. So setting up some quick agents in frameworks like \",/*#__PURE__*/e(i,{href:\"https://www.llamaindex.ai/\",nodeId:\"a3ufc1L7f\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Llama index\"})}),\" help with data organization and transformation. There are also a number of low/no code tools like \",/*#__PURE__*/e(i,{href:\"https://retool.com/\",nodeId:\"a3ufc1L7f\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Retool\"})}),\" that can help a PM quickly mock up ideas to discuss with stakeholders. As AI is interactive, it\u2019s not like the old days of getting a screen mockup to share, these things need interactivity.\\xa0\"]}),/*#__PURE__*/t(\"p\",{children:[\"Of course, traditional product management tools, when augmented with AI capabilities, can become even more powerful. In my daily workflow I rely on AI tools like \",/*#__PURE__*/e(i,{href:\"https://chatgpt.com/\",nodeId:\"a3ufc1L7f\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"ChatGPT\"})}),\", \",/*#__PURE__*/e(i,{href:\"https://claude.ai/login?returnTo=%2F%3F\",nodeId:\"a3ufc1L7f\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Claude\"})}),\", \",/*#__PURE__*/e(i,{href:\"https://www.perplexity.ai/\",nodeId:\"a3ufc1L7f\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Perplexity\"})}),\", and our own tool, \",/*#__PURE__*/e(i,{href:\"https://artium.ai/apex\",nodeId:\"a3ufc1L7f\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"APEX\"})}),\". A lot of PMs use tools like \",/*#__PURE__*/e(i,{href:\"https://www.figma.com/figjam/\",nodeId:\"a3ufc1L7f\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"FigJam\"})}),\" and \",/*#__PURE__*/e(i,{href:\"https://miro.com/?gclsrc=aw.ds&utm_source=google&utm_medium=cpc&utm_campaign=S%7CGOO%7CBRN%7CUS%7CEN-EN%7CBrand%7CExact&utm_adgroup=&adgroupid=140324301065&utm_custom=18258206285&utm_content=668037264392&utm_term=miro&matchtype=e&device=c&location=9030961&gad_source=1&gclid=CjwKCAjw5Ky1BhAgEiwA5jGujjDPJVs5vnIYFfB9P83xx_7lMYq5H5Zvgq0n9AEkar3aHpxsji9F4BoCDDMQAvD_BwE\",nodeId:\"a3ufc1L7f\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"Miro\"})}),\" to create maps of complex ideas and consolidate feedback. These tools, augmented with AI, provide a great path for getting from scattered info to a honed message and strong direction.\\xa0\"]}),/*#__PURE__*/e(\"p\",{children:\"LLM tools assist with things like generating user stories, analyzing data, and providing actionable insights. It\u2019s like I have a buddy now where I can just write a line that says, \u201CHey, I want to do this\u201D and the agent will ask me 3 or 4 questions, and then spit out an entire user story.\\xa0\"}),/*#__PURE__*/e(\"p\",{children:\"The killer tools are yet to come.\"}),/*#__PURE__*/e(\"ol\",{start:\"4\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"h6\",children:/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"How do you stay updated with the latest advancements in AI and ensure that your team is equipped with cutting-edge knowledge?\"})})})}),/*#__PURE__*/e(\"p\",{children:\"I have a bunch of AI agents that I've created that can distill down the firehose of information that comes in around AI. And so what I do is I look for new research papers that have come out and pass it through an AI, and just share the summary with the link to the paper. So I am up-to-date with any significant update and changes that are going on and always share those with my team.\"}),/*#__PURE__*/t(\"p\",{children:[\"There are also a few popular Youtube channels and newsletters that I subscribe to that bring me weekly news or summaries. Linkedin is also a great resource. Also I never stop taking courses online from reliable sources like \",/*#__PURE__*/e(i,{href:\"http://deeplearing.ai/\",nodeId:\"a3ufc1L7f\",openInNewTab:!1,smoothScroll:!1,children:/*#__PURE__*/e(\"a\",{children:\"deeplearing.ai \"})}),\"- learn by doing is great, but even better mixed with study.\"]}),/*#__PURE__*/e(\"ol\",{start:\"5\",children:/*#__PURE__*/e(\"li\",{\"data-preset-tag\":\"h6\",children:/*#__PURE__*/e(\"h6\",{children:/*#__PURE__*/e(\"strong\",{children:\"How do you foresee the role of AI in product development evolving in the next 5-10 years?\"})})})}),/*#__PURE__*/e(\"p\",{children:\"There's a lot that can happen in 5 years. But I think that it's like most jobs, it's going to continue to be redefined. When you're collaborating with an AI it will become smarter and will get to know you better, and with that it will be able to do more and more of the work.\\xa0\"}),/*#__PURE__*/e(\"p\",{children:\"Then the question is\u2026 What do we want to do with the time that AI can help free up? And I think that differs from person to person. You can put that effort into research and looking into the future and seeing what that product will become, or you can put that effort into being more strategic with your stakeholders and so on. So I think overall what AI is going to do is free up a lot of time for product people to be more strategic and level up their careers.\"})]});\nexport const __FramerMetadata__ = {\"exports\":{\"richText12\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"richText2\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"richText1\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"richText6\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"richText5\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"richText10\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"richText4\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"richText\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"richText8\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"richText7\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"richText9\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"richText11\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"richText3\":{\"type\":\"variable\",\"annotations\":{\"framerContractVersion\":\"1\"}},\"__FramerMetadata__\":{\"type\":\"variable\"}}}"],
  "mappings": "uUAAyS,IAAMA,EAAsBC,EAAIC,EAAS,CAAC,SAAS,CAAcC,EAAE,IAAI,CAAC,SAAS,+vBAA0vB,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,mBAAmB,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,28BAA47B,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4cAAuc,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,MAAM,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4oBAAuoB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,odAAod,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,wBAAwB,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,mjBAAmjB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6XAA6X,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,sBAAsB,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,kLAAkL,CAAC,EAAeF,EAAE,KAAK,CAAC,SAAS,CAAcE,EAAE,KAAK,CAAC,kBAAkB,IAAI,MAAM,CAAC,qBAAqB,OAAO,sBAAsB,eAAe,2BAA2B,MAAM,EAAE,SAAsBA,EAAE,IAAI,CAAC,SAAS,yCAAyC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,MAAM,CAAC,qBAAqB,OAAO,sBAAsB,eAAe,2BAA2B,MAAM,EAAE,SAAsBA,EAAE,IAAI,CAAC,SAAS,0CAA0C,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,MAAM,CAAC,qBAAqB,OAAO,sBAAsB,eAAe,2BAA2B,MAAM,EAAE,SAAsBA,EAAE,IAAI,CAAC,SAAS,6BAA6B,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,ymBAAymB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4VAAuV,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,2BAA2B,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,sNAAsN,CAAC,EAAeF,EAAE,KAAK,CAAC,SAAS,CAAcE,EAAE,KAAK,CAAC,kBAAkB,IAAI,MAAM,CAAC,qBAAqB,OAAO,sBAAsB,eAAe,2BAA2B,MAAM,EAAE,SAAsBA,EAAE,IAAI,CAAC,SAAS,0DAA0D,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,MAAM,CAAC,qBAAqB,OAAO,sBAAsB,eAAe,2BAA2B,MAAM,EAAE,SAAsBA,EAAE,IAAI,CAAC,SAAS,oDAAoD,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,MAAM,CAAC,qBAAqB,OAAO,sBAAsB,eAAe,2BAA2B,MAAM,EAAE,SAAsBA,EAAE,IAAI,CAAC,SAAS,wEAAwE,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,kSAAkS,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,wBAAwB,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,wDAAwD,CAAC,EAAeF,EAAE,KAAK,CAAC,SAAS,CAAcE,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,uEAAuE,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,sEAAsE,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,+DAA+D,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,sDAAsD,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,kBAAkB,IAAI,SAAsBA,EAAE,IAAI,CAAC,SAAS,2DAA2D,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeF,EAAE,KAAK,CAAC,SAAS,CAAcE,EAAE,SAAS,CAAC,SAAS,gFAA2E,CAAC,EAAE,QAAG,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,IAAI,MAAM,CAAC,qBAAqB,OAAO,sBAAsB,eAAe,2BAA2B,MAAM,EAAE,SAAsBF,EAAE,IAAI,CAAC,SAAS,CAAC,uBAA+BE,EAAEC,EAAE,CAAC,KAAK,yDAAyD,OAAO,YAAY,aAAa,GAAG,aAAa,GAAG,SAAsBD,EAAE,IAAI,CAAC,SAAS,gCAAgC,CAAC,CAAC,CAAC,EAAE,iDAAiD,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,IAAI,MAAM,CAAC,qBAAqB,OAAO,sBAAsB,eAAe,2BAA2B,MAAM,EAAE,SAAsBF,EAAE,IAAI,CAAC,SAAS,CAAC,oEAAiFE,EAAEC,EAAE,CAAC,KAAK,2GAA2G,OAAO,YAAY,aAAa,GAAG,aAAa,GAAG,SAAsBD,EAAE,IAAI,CAAC,SAAS,8BAA8B,CAAC,CAAC,CAAC,EAAE,iHAAiH,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,IAAI,MAAM,CAAC,qBAAqB,OAAO,sBAAsB,eAAe,2BAA2B,MAAM,EAAE,SAAsBA,EAAE,IAAI,CAAC,SAAS,6TAA6T,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oXAAoX,CAAC,CAAC,CAAC,CAAC,EAAeE,EAAuBJ,EAAIC,EAAS,CAAC,SAAS,CAAcC,EAAE,IAAI,CAAC,SAAS,+TAA+T,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,oEAAoE,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,qfAAie,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,k3BAAw2B,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,imBAAulB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,ipBAA4oB,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,uHAAuH,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,s5BAAu4B,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,6EAA6E,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,isBAAkrB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,wgBAA8f,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,iEAAiE,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6dAAmd,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,8jBAA8jB,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,yEAAyE,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,+ZAA0Z,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,gZAAkX,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,qgBAAif,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,+QAA+Q,CAAC,CAAC,CAAC,CAAC,EAAeG,EAAuBL,EAAIC,EAAS,CAAC,SAAS,CAAcC,EAAE,IAAI,CAAC,SAAS,wWAA8V,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,wFAAwF,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,+ZAAqZ,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,mXAAmX,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,6FAA0GE,EAAEC,EAAE,CAAC,KAAK,+EAA+E,OAAO,YAAY,aAAa,GAAG,aAAa,GAAG,SAAsBD,EAAE,IAAI,CAAC,SAAS,YAAY,CAAC,CAAC,CAAC,EAAE,whBAAqf,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,gIAA2H,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,0FAA0F,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,whBAAmhB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,gsBAAmrB,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,iDAAiD,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oVAA+U,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0YAAiX,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4NAA4N,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,4LAAuL,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,kfAA8d,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,umBAA8kB,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,uMAAuM,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,mdAAyc,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,gmBAAukB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,sbAA4a,CAAC,CAAC,CAAC,CAAC,EAAeI,EAAuBN,EAAIC,EAAS,CAAC,SAAS,CAAcD,EAAE,IAAI,CAAC,SAAS,CAAC,gJAA6JE,EAAE,SAAS,CAAC,SAAS,oBAAoB,CAAC,EAAE,8KAA8K,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,2NAA2N,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,gJAAgJ,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,0EAA0E,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4NAAuN,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,obAAob,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,ugBAAkgB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,kLAAkL,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,uFAAuF,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,ydAA+c,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,mcAA8b,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4OAAuO,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,mGAAmG,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,uKAAuK,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,uTAAuT,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,yTAAyT,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0hBAA0hB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,ubAA8Z,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,+JAA+J,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oRAAoR,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6VAAwV,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,kIAAkI,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6TAA6T,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,qTAAgT,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,saAA4Z,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,2IAAsI,CAAC,CAAC,CAAC,CAAC,EAAeK,EAAuBL,EAAID,EAAS,CAAC,SAAsBC,EAAE,KAAK,CAAC,SAAS,6FAA6F,CAAC,CAAC,CAAC,EAAeM,EAAuBR,EAAIC,EAAS,CAAC,SAAS,CAAcC,EAAE,IAAI,CAAC,SAAS,0JAAqJ,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,+MAA+M,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,uLAAuL,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,0FAA0F,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,iXAA4W,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,mkBAAmkB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,mOAAmO,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0RAAgR,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,uFAAkF,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,45BAAm5B,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,gjBAAgjB,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,gFAAgF,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,qWAAgW,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,61BAAo0B,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,4GAA4G,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,whBAAmhB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4dAAkd,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,6vBAAgwBE,EAAEC,EAAE,CAAC,KAAK,6BAA6B,OAAO,YAAY,aAAa,GAAG,aAAa,GAAG,SAAsBD,EAAE,IAAI,CAAC,SAAS,WAAW,CAAC,CAAC,CAAC,EAAE,KAAkBA,EAAEC,EAAE,CAAC,KAAK,0BAA0B,OAAO,YAAY,aAAa,GAAG,aAAa,GAAG,SAAsBD,EAAE,IAAI,CAAC,SAAS,QAAQ,CAAC,CAAC,CAAC,EAAE,KAAkBA,EAAEC,EAAE,CAAC,KAAK,6BAA6B,OAAO,YAAY,aAAa,GAAG,aAAa,GAAG,SAAsBD,EAAE,IAAI,CAAC,SAAS,YAAY,CAAC,CAAC,CAAC,EAAE,wBAAqCA,EAAEC,EAAE,CAAC,KAAK,uEAAuE,OAAO,YAAY,aAAa,GAAG,aAAa,GAAG,SAAsBD,EAAE,IAAI,CAAC,SAAS,eAAe,CAAC,CAAC,CAAC,EAAE,ySAAyS,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeO,EAAuBP,EAAID,EAAS,CAAC,SAAsBC,EAAE,KAAK,CAAC,SAAS,oEAAoE,CAAC,CAAC,CAAC,EAAeQ,EAAuBV,EAAIC,EAAS,CAAC,SAAS,CAAcC,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,+MAA+M,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,ggBAAif,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,8VAA8V,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6PAAmP,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,qKAAqK,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,uHAAuH,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,gOAAgO,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,mPAA8O,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,mhCAA0/B,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,4IAA4I,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4lBAAulB,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,2NAA2N,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,2vBAAivB,CAAC,CAAC,CAAC,CAAC,EAAeS,EAAuBT,EAAID,EAAS,CAAC,SAAsBC,EAAE,KAAK,CAAC,SAAS,8FAA8F,CAAC,CAAC,CAAC,EAAeU,EAAuBZ,EAAIC,EAAS,CAAC,SAAS,CAAcD,EAAE,IAAI,CAAC,SAAS,CAAC,sNAA8NE,EAAEC,EAAE,CAAC,KAAK,+CAA+C,OAAO,YAAY,aAAa,GAAG,aAAa,GAAG,SAAsBD,EAAE,IAAI,CAAC,SAAS,kBAAkB,CAAC,CAAC,CAAC,EAAE,mQAA8P,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,mEAAmE,CAAC,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAE,SAAS,CAAC,SAAS,IAAI,CAAC,EAAE,+jBAA0jB,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,iHAAiH,CAAC,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAE,SAAS,CAAC,SAAS,KAAK,CAAC,EAAE,8DAA8D,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4OAA4O,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oOAAoO,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,oGAAoG,CAAC,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAE,SAAS,CAAC,SAAS,IAAI,CAAC,EAAE,4lBAAulB,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,mHAAmH,CAAC,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAE,SAAS,CAAC,SAAS,IAAI,CAAC,EAAE,ggBAAsf,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oVAA+U,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,qFAAqF,CAAC,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAE,SAAS,CAAC,SAAS,IAAI,CAAC,EAAE,kMAA6L,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,yNAAyN,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,yGAAyG,CAAC,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAcE,EAAE,SAAS,CAAC,SAAS,IAAI,CAAC,EAAE,gRAAsQ,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,wYAA8X,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,2FAAsF,CAAC,CAAC,CAAC,CAAC,EAAeW,EAAwBb,EAAIC,EAAS,CAAC,SAAS,CAAcC,EAAE,IAAI,CAAC,SAAS,mWAAmW,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4WAA4W,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,yBAAsCE,EAAEC,EAAE,CAAC,KAAK,yCAAyC,OAAO,YAAY,aAAa,GAAG,aAAa,GAAG,SAAsBD,EAAE,IAAI,CAAC,SAAS,aAAa,CAAC,CAAC,CAAC,EAAE,uNAAuN,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,sTAAsT,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0NAA0N,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,wCAAwC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,i3BAA42B,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,4xBAA4xB,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,sXAAsX,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,0XAA0X,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6RAA6R,CAAC,EAAeA,EAAE,KAAK,CAAC,SAAS,uDAAuD,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,kDAAkD,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,qSAAqS,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,wIAAwI,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,oDAAoD,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,wJAAwJ,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,kGAAkG,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,wDAAwD,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6QAA6Q,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,yHAAyH,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,8RAA8R,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,6BAA6B,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,kFAAkF,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,uDAAuD,CAAC,CAAC,CAAC,EAAeA,EAAEY,EAAE,IAAI,CAAC,UAAU,qBAAqB,MAAM,CAAC,iBAAiB,YAAY,YAAY,sBAAsB,OAAO,OAAO,MAAM,MAAM,EAAE,SAAsBZ,EAAEa,EAAE,CAAC,oBAAoB,sEAAsE,SAASC,GAAgBd,EAAEe,EAAE,CAAC,GAAGD,EAAE,KAAK,MAAM,WAAW,GAAG,UAAU,iBAAiB,IAAI,oDAAoD,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeE,EAAwBhB,EAAID,EAAS,CAAC,SAAsBC,EAAE,KAAK,CAAC,SAAS,4DAA4D,CAAC,CAAC,CAAC,EAAeiB,EAAwBnB,EAAIC,EAAS,CAAC,SAAS,CAAcC,EAAE,KAAK,CAAC,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,KAAK,SAAsBA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,iIAAiI,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,6VAA6V,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oQAAoQ,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oUAAoU,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,IAAI,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,KAAK,SAAsBA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,gFAAgF,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,kNAAkN,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,sZAAsZ,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,qVAAqV,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,keAAke,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,8iBAA8iB,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,8CAA2DE,EAAEC,EAAE,CAAC,KAAK,yBAAyB,OAAO,YAAY,aAAa,GAAG,aAAa,GAAG,SAAsBD,EAAE,IAAI,CAAC,SAAS,MAAM,CAAC,CAAC,CAAC,EAAE,qQAAqQ,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,IAAI,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,KAAK,SAAsBA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,iHAAiH,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,8NAA2OE,EAAEC,EAAE,CAAC,KAAK,6BAA6B,OAAO,YAAY,aAAa,GAAG,aAAa,GAAG,SAAsBD,EAAE,IAAI,CAAC,SAAS,aAAa,CAAC,CAAC,CAAC,EAAE,sGAAmHA,EAAEC,EAAE,CAAC,KAAK,sBAAsB,OAAO,YAAY,aAAa,GAAG,aAAa,GAAG,SAAsBD,EAAE,IAAI,CAAC,SAAS,QAAQ,CAAC,CAAC,CAAC,EAAE,yMAAoM,CAAC,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,qKAAkLE,EAAEC,EAAE,CAAC,KAAK,uBAAuB,OAAO,YAAY,aAAa,GAAG,aAAa,GAAG,SAAsBD,EAAE,IAAI,CAAC,SAAS,SAAS,CAAC,CAAC,CAAC,EAAE,KAAkBA,EAAEC,EAAE,CAAC,KAAK,0CAA0C,OAAO,YAAY,aAAa,GAAG,aAAa,GAAG,SAAsBD,EAAE,IAAI,CAAC,SAAS,QAAQ,CAAC,CAAC,CAAC,EAAE,KAAkBA,EAAEC,EAAE,CAAC,KAAK,6BAA6B,OAAO,YAAY,aAAa,GAAG,aAAa,GAAG,SAAsBD,EAAE,IAAI,CAAC,SAAS,YAAY,CAAC,CAAC,CAAC,EAAE,uBAAoCA,EAAEC,EAAE,CAAC,KAAK,yBAAyB,OAAO,YAAY,aAAa,GAAG,aAAa,GAAG,SAAsBD,EAAE,IAAI,CAAC,SAAS,MAAM,CAAC,CAAC,CAAC,EAAE,iCAA8CA,EAAEC,EAAE,CAAC,KAAK,gCAAgC,OAAO,YAAY,aAAa,GAAG,aAAa,GAAG,SAAsBD,EAAE,IAAI,CAAC,SAAS,QAAQ,CAAC,CAAC,CAAC,EAAE,QAAqBA,EAAEC,EAAE,CAAC,KAAK,iXAAiX,OAAO,YAAY,aAAa,GAAG,aAAa,GAAG,SAAsBD,EAAE,IAAI,CAAC,SAAS,MAAM,CAAC,CAAC,CAAC,EAAE,8LAA8L,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,qTAAsS,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,mCAAmC,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,IAAI,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,KAAK,SAAsBA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,+HAA+H,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,oYAAoY,CAAC,EAAeF,EAAE,IAAI,CAAC,SAAS,CAAC,mOAAgPE,EAAEC,EAAE,CAAC,KAAK,yBAAyB,OAAO,YAAY,aAAa,GAAG,aAAa,GAAG,SAAsBD,EAAE,IAAI,CAAC,SAAS,iBAAiB,CAAC,CAAC,CAAC,EAAE,8DAA8D,CAAC,CAAC,EAAeA,EAAE,KAAK,CAAC,MAAM,IAAI,SAAsBA,EAAE,KAAK,CAAC,kBAAkB,KAAK,SAAsBA,EAAE,KAAK,CAAC,SAAsBA,EAAE,SAAS,CAAC,SAAS,2FAA2F,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,yRAAyR,CAAC,EAAeA,EAAE,IAAI,CAAC,SAAS,odAA+c,CAAC,CAAC,CAAC,CAAC,EACl6yEkB,EAAqB,CAAC,QAAU,CAAC,WAAa,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,UAAY,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,UAAY,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,UAAY,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,UAAY,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,WAAa,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,UAAY,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,SAAW,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,UAAY,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,UAAY,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,UAAY,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,WAAa,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,UAAY,CAAC,KAAO,WAAW,YAAc,CAAC,sBAAwB,GAAG,CAAC,EAAE,mBAAqB,CAAC,KAAO,UAAU,CAAC,CAAC",
  "names": ["richText", "u", "x", "p", "Link", "richText1", "richText2", "richText3", "richText4", "richText5", "richText6", "richText7", "richText8", "richText9", "richText10", "motion", "ComponentPresetsConsumer", "t", "Youtube", "richText11", "richText12", "__FramerMetadata__"]
}