Blog

This article continues our series on harnessing LLMs for real-world product development through our Mission Control project. As we've established throughout this series, building production-grade software demands more than flashy demos—it requires deliberate architectural decisions that stand up to real-world demands.Two of our four North Star strategic goals directly shaped every architectural choice we made.&nbsp;<ul><li style="">First, we needed an architecture that delivers Reliability, Performance, and Scalability from day one.&nbsp; </li><li style="">Second, our tech stack had to enable Flexibility and Extensibility as our product evolves.&nbsp;</li></ul>These weren't abstract principles—they were concrete requirements that guided every decision from backend infrastructure to frontend frameworks.<blockquote>The stakes were clear: if Mission Control couldn't demonstrate these capabilities, it would fail as a genuine test of real-world product development. Our architectural planning process needed to prove that LLMs could accelerate not just code generation, but the strategic thinking that creates lasting technical foundations.</blockquote><h3>The Decision Framework: From Breadth to Depth </h3>Our methodology followed the same decision framework we've refined throughout this series: systematic "shootouts" between multiple LLM models using identical prompts to research architectural options. This wasn't random experimentation—it was structured evaluation designed to surface the best solutions quickly.The process began with broad research prompts sent simultaneously to <a data-mce-href="https://claude.ai/" href="https://claude.ai/" title="Claude AI" target="_blank" rel="noopener">Claude</a>, <a data-mce-href="https://chatgpt.com/" href="https://chatgpt.com/" title="ChatGPT AI" target="_blank" rel="noopener">ChatGPT</a>, <a data-mce-href="https://copilot.microsoft.com/" href="https://copilot.microsoft.com/" title="Microsoft CoPilot" target="_blank" rel="noopener">CoPilot</a>, and <a data-mce-href="https://www.perplexity.ai/" href="https://www.perplexity.ai/" title="Perplexity AI" target="_blank" rel="noopener">Perplexity</a>. Each LLM received the same context: our product requirements, performance expectations, and scalability needs. The goal was maximum coverage of potential solutions in minimum time.From there, we moved through an iterative breadth-to-depth process. We narrowed prompts to focus on the most promising options while eliminating LLMs that consistently delivered weaker responses. This natural selection approach quickly identified which models excelled at architectural analysis versus surface-level suggestions.The framework ultimately narrowed to two primary LLM chat threads conducting deep research across architectural options. Each generated detailed decision documents that became our evaluation foundation. These documents included five critical components:<ul><li>Technology Capability Assessment: What each component could actually do, not marketing promises but real-world functionality based on documentation and case studies. </li><li>Mission Control Alignment: How each option supported our specific product requirements, from user authentication to content management to B2B SaaS capabilities. </li><li>Multi-Criteria Analysis: Pros and cons evaluated against selection criteria including costs, extensibility, performance, robustness of capabilities, and integration complexity. </li><li>Risk Assessment: Potential failure points and mitigation strategies, because every architectural decision carries technical debt implications. </li><li>Community and Documentation Evaluation: The strength of community support and quality of documentation—a factor that would prove crucial for LLM productivity gains later in development. </li></ul>This systematic approach transformed what could have been weeks of research into days of focused analysis. The LLMs didn't just accelerate information gathering—they structured it for better decision-making.<h3>Critical Architecture Decisions </h3>Our research process required decisions across five major architectural domains: backend infrastructure, database technology, authentication systems, service layer architecture, and frontend frameworks. Each decision interconnected with the others, creating a web of technical dependencies that demanded careful consideration.<ul><li>Backend Infrastructure: Evaluated AWS, GCP/Firebase, and Azure, focusing on startup needs like limited budgets, small teams, rapid iteration, and future scalability.&nbsp; </li><li>Database Decisions: Considered relational, document, and hybrid databases, as well as deployment options like managed services, self-hosted, and containerized solutions. Modeled scenarios from launch to scaling with LLMs.&nbsp; </li><li>Authentication and User Management: Explored built-in solutions, third-party services, and hybrid approaches to balance robust user experiences with strong security.&nbsp; </li><li>Service Layer: Assessed REST APIs, GraphQL, and hybrid options to weigh performance, flexibility, and development complexity.&nbsp; </li><li>Frontend Frameworks: Balanced short-term productivity with long-term goals, including potential expansion to mobile, desktop, and new platforms. </li></ul>Throughout this process, our third North Star goal remained paramount: we needed to build something that demonstrated Real-World, Universal Capabilities. Every product team faces the same core requirements—user authentication, profile management, roles and permissions, content management, search and tagging, and B2B SaaS features like customer data segregation.Our tech stack choices had to support all these capabilities without creating architectural bottlenecks. This wasn't about building the minimum viable product—it was about creating a foundation that could evolve from MVP to enterprise-grade platform.<h3>The Winning Tech Stack </h3>After extensive research and evaluation, our analysis yielded a specific technology combination: GCP/Firebase for backend infrastructure, containerized PostgreSQL for the database, Directus as our headless CMS, Hasura for the API layer, and Flutter for the frontend framework.These choices weren't made in isolation—each component strengthened the others while addressing our strategic goals.<a data-mce-href="https://cloud.google.com/" href="https://cloud.google.com/" title="Google Cloud Home" target="_blank" rel="noopener">Google Cloud</a> / <a data-mce-href="https://firebase.google.com/" href="https://firebase.google.com/" title="Firebase Home" target="_blank" rel="noopener">Firebase</a> emerged as our backend infrastructure winner for several compelling reasons. The Firebase ecosystem provides cost-effective document-based data management with authentication and user management built in. Google Analytics 4 integration gives us powerful analytics capabilities without additional complexity. Storage capabilities for assets and messaging systems come standard.The cost structure perfectly aligns with startup realities—essentially free until significant traction develops, then backed by GCP's enterprise-grade scalability for future growth. The usability factor proved decisive; Firebase enables rapid development without sacrificing professional capabilities. Google's backing ensures long-term viability, while robust documentation and community support maximize LLM productivity gains.<a data-mce-href="https://www.postgresql.org/" href="https://www.postgresql.org/" title="PostgreSQL" target="_blank" rel="noopener">PostgreSQL</a> represented the safe, scalable choice for our primary database. As a battle-tested relational database with excellent performance characteristics and virtually unlimited scalability, PostgreSQL offers the reliability our first North Star goal demanded. The extensive community support and documentation also aligned with our LLM productivity requirements.<a data-mce-href="https://directus.io/" href="https://directus.io/" title="Direcus Home" target="_blank" rel="noopener">Directus</a> solved our headless CMS needs with robust content management capabilities and cost-effective pricing. Among all CMS technologies we evaluated, Directus offered the best balance of functionality and affordability. Its integration capabilities with Firebase/GCP and Flutter sealed the decision, enabling us to demonstrate a complete content pipeline within Mission Control.<a data-mce-href="https://hasura.io/" href="https://hasura.io/" title="Hasura Home" target="_blank" rel="noopener">Hasura</a> provided our API layer solution with strong PostgreSQL and Firebase integration. The GraphQL flexibility it offers enables efficient frontend queries while maintaining backend performance. Hasura's approach to API generation dramatically reduces boilerplate code while providing the extensibility our second North Star goal required.<a data-mce-href="https://flutter.dev/" href="https://flutter.dev/" title="Flutter Development Home" target="_blank" rel="noopener">Flutter</a> won our frontend framework evaluation as a cross-platform solution that enables immediate web app development while preserving mobile expansion options. Flutter's growing ecosystem, strong Firebase/GCP integration, and compatibility with our other tech stack components made it the clear choice. The ability to reach iOS, Android, and web platforms from a single codebase offers exactly the flexibility and extensibility we needed.<h3>The Community Support Advantage </h3>One crucial insight emerged from our architectural planning process: when building systems you plan to enhance with LLM intelligence, choose well-supported, well-documented technologies. This principle became a key selection criterion that influenced every decision.LLMs are only as capable as the training data they've learned from. Technologies with extensive documentation, active communities, and widespread adoption generate more training examples, leading to better LLM responses and higher productivity gains during development.This insight proved prescient during actual development. Throughout the Mission Control build process, I received more consistent and substantive assistance from LLMs when working with Firebase, PostgreSQL, Hasura, and Flutter. These technologies have extensive documentation and active communities, resulting in training data that enables LLMs to provide detailed, accurate, and contextually appropriate guidance.Directus presented the exception that proved the rule. As the least supported component in our tech stack, Directus consistently generated weaker LLM responses. Documentation gaps and smaller community adoption meant fewer high-quality examples for LLMs to learn from. This translated to more trial-and-error development, more debugging time, and lower overall productivity gains.The lesson is clear: architectural decisions should factor in LLM ecosystem support as a legitimate selection criterion. In an AI-accelerated development world, choosing technologies that maximize LLM effectiveness becomes a competitive advantage.<h3>Results and Recommendations </h3>The process of using LLMs (large language models) to assist in architectural planning and tech stack selection delivered impressive results. By accelerating research and decision-making, we achieved significant productivity gains and created a strong, scalable foundation for development. Here are the key recommendations based on our experience:<ul><li>Accelerate research and evaluation: Use LLMs to speed up initial research, compare options, analyze criteria, and make informed choices faster.&nbsp; </li><li>Structure decision-making: Adopt a systematic framework to evaluate risks, integration challenges, and scalability considerations upfront rather than encountering surprises later.&nbsp; </li><li>Leverage LLMs as partners, not decision-makers: Use LLMs to gather data, structure comparisons, and highlight trade-offs, but keep final strategic decisions aligned with your specific goals in human hands.&nbsp; </li><li>Build for scalability and adaptability: Prioritize architectural decisions that support long-term growth, adaptability, and extensibility under real-world pressures.&nbsp; </li><li>Focus on foundations: A solid tech stack and well-thought-out architecture make everything built on top more reliable, scalable, and successful. </li></ul>By combining the speed and insights of AI with human strategic judgment, teams can make better architectural decisions in less time. This approach not only accelerates development but also ensures the foundation is strong enough to support future growth and innovation.Next up: we'll dive into how LLM assistance boosted the productivity of the development process and what we learned along the way to maximize results and avoid rollbacks.&nbsp;

How AI Boosted the Architectural Design Process for Mission Control

This article is part of our ongoing series on harnessing LLMs to deliver real-world value in product development. Through the Mission Control project, we aim to provide a transparent view of how AI accelerates high-performing product teams. My core strengths are in product management, leadership, and some engineering, but UX design has always been an area where I’ve relied on seasoned partners to turn vision into meaningful user experiences.With Mission Control, I intentionally worked solo—a deliberate test of how much productivity could be achieved by leveraging an LLM as a dedicated UX design assistant. <blockquote>The objective was clear: could AI efficiently bridge my expertise gap and deliver a production-ready design?</blockquote><h3>The AI Design Shootout: From Prompt to Prototype </h3>Every product design project needs a solid foundation—the product requirements document (PRD). Generated with the support of <a data-mce-href="https://www.chatprd.ai/" href="https://www.chatprd.ai/" title="ChatPRD AI" target="_blank" rel="noopener">ChatPRD</a>, our PRD captured strategic statements on the core problem, solution, vision, high-level requirements, and essential features. This served as our single source of truth for what needed to be built.With the requirements in place, I kicked off a “shootout” among leading AI tools. I fed the PRD into <a data-mce-href="https://claude.ai/" href="https://claude.ai/" title="Claude AI" target="_blank" rel="noopener">Claude</a>, <a data-mce-href="https://copilot.microsoft.com/" href="https://copilot.microsoft.com/" title="Microsoft CoPilot" target="_blank" rel="noopener">Microsoft CoPilot</a>, <a data-mce-href="https://chatgpt.com/" href="https://chatgpt.com/" title="ChatGPT" target="_blank" rel="noopener">ChatGPT</a>, and <a data-mce-href="https://www.figma.com/ux-design-tool/" href="https://www.figma.com/ux-design-tool/" title="Figma UX Design" target="_blank" rel="noopener">Figma’s</a> AI-driven prototyping features, asking each to generate a visual concept for the Mission Control dashboard. This let me objectively compare how each platform translated complex product needs into coherent, actionable interfaces.<h3>Iteration and Refinement: Finding the Winning Workflow </h3>The initial concepts from each platform were starting points. Real progress began with focused, iterative cycles—review, critique, revised prompts, and new outputs. The differences in output quality and practical utility quickly became clear.Through this process, I narrowed my toolkit to <a data-mce-href="https://claude.ai/" href="https://claude.ai/" title="Claude AI" target="_blank" rel="noopener">Claude</a> and <a data-mce-href="https://www.figma.com/ux-design-tool/" href="https://www.figma.com/ux-design-tool/" title="Figma UX Design Tool" target="_blank" rel="noopener">Figma</a>.Claude consistently generated mockups with visual impact and efficiency, aligning with the Mission Control audience and requirements. Its workflows made it possible to go from requirements to high-fidelity ideas quickly, without compromising on clarity.Figma, meanwhile, was indispensable for structure and version control. While Claude drove creative output, Figma provided the rigor required to organize and manage design iterations. The resulting workflow was a two-step process:<ol><li>Generate top-level visual concepts and mockups in Claude. </li><li>Organize, refine, and version those assets in Figma. </li></ol>This “1-2 UXD” approach combined AI’s creative speed with the reliability and professionalism of a best-in-class design platform.<figure id="figure-o1opkv3stpmesogwjy" data-full-width=""><img id="o1opkv3stpmesogwjy" src="https://cdn.durable.co/blocks/12pOuph92UT6xeKZHd5mxQ03m212sFQQTWRuAkBVS524d2lAD4MbKfSJgPDC2sWc.png" alt="claude prompts to figma designs" title="claude prompts to figma designs" class="object-cover bg-gray-100 rounded-sm md:rounded-md lg:rounded-lg" style="aspect-ratio: 16/9; object-position: 0% 50%;" data-corners="default" data-aspect="16:9" data-caption="Claude Prompts to Figma Designs" data-positionx="0%" data-positiony="50%" data-fullwidth="true" data-mce-src="https://cdn.durable.co/blocks/12pOuph92UT6xeKZHd5mxQ03m212sFQQTWRuAkBVS524d2lAD4MbKfSJgPDC2sWc.png" data-mce-style="aspect-ratio: 16/9; object-position: 0% 50%;"><figcaption class="body-normal text-gray-500 mt-2">Claude Prompts to Figma Designs</figcaption></figure><h3>From Design to Code: The Implementation Challenge </h3>With finalized designs in Figma, the next challenge was implementation: translating those designs into robust Dart/Flutter code for our front end. This is often a critical pain point, so I evaluated two main approaches.<ul><li>Use an automated exporter to turn Figma designs directly into Dart code. While this promises speed, the downside is often messy, unoptimized code that hampers scalability and maintenance. Work with VS Code Copilot, prompting it with targeted screen grabs of UI components from Figma. By generating code for each widget—cards, buttons, nav bars—individually, the result is far cleaner. </li><li>Building out the UI widget by widget produced a scalable and well-structured codebase, aligning with Flutter best practices. Although more hands-on, this manual, AI-assisted approach set a high standard for implementation quality and long-term maintainability.</li></ul><figure id="figure-dbb5kgntjgmesoknwk" data-full-width=""><img id="dbb5kgntjgmesoknwk" src="https://cdn.durable.co/blocks/29f2XSVImigvWgZTyByDsRn6YbmLgW0vWnTFgyHmMwAOTZfLodYqcD2eUNcK0fhh.png" alt="flutter widget implementation" title="flutter widget implementation" class="object-cover bg-gray-100 rounded-sm md:rounded-md lg:rounded-lg" style="aspect-ratio: 16/9; object-position: 50% 50%;" data-corners="default" data-aspect="16:9" data-caption="Flutter Widget-by-Widget Implementation " data-positionx="50%" data-positiony="50%" data-fullwidth="true" data-mce-src="https://cdn.durable.co/blocks/29f2XSVImigvWgZTyByDsRn6YbmLgW0vWnTFgyHmMwAOTZfLodYqcD2eUNcK0fhh.png" data-mce-style="aspect-ratio: 16/9; object-position: 50% 50%;"><figcaption class="body-normal text-gray-500 mt-2">Flutter Widget-by-Widget Implementation</figcaption></figure><h3>The Results: A 2–5x Gain in Design Efficiency </h3>The final workflow took us from product requirements, through interactive mockups, to a production-ready front end. Leveraging LLMs as a UX design partner, I saw a 2–5x improvement in efficiency compared to tackling the process without AI support.This system enabled me to deliver a user-centric design that fully supported the project’s objectives, despite not having formal UX training. Here, AI served as a true force multiplier, transforming product vision into validated design at speed.<h3>Process Improvements for Next Time </h3>Is this approach a “silver bullet” for UX design? Not quite. Mission Control was a time-boxed effort, so I was limited in how far I could refine process and tooling. That said, I see clear paths for optimization in future engagements:<ul><li>Explore more advanced automation and Figma-Futter integrations (e.g., FlutterFlow, Rocket, FlutterViz) to streamline the coding of individual frontend components. </li><li>Bring experienced UX design professionals into the AI-augmented workflow. My productivity gains were limited by my own gaps. With expert oversight, I’d expect results closer to 5–20x improvements—truly transformative. </li><li>Strengthen our Figma asset management with versioning tools and tighter integration into source control, enhancing team workflows and design traceability. </li></ul><h3>Conclusion: Expertise Defines the Exponential Gain </h3>This experience reinforces what we see again and again: expertise shapes the ROI of LLM-powered product development. When applied in areas of deep knowledge—like product strategy or backend architecture—AI can enable exponential acceleration.In UX design, where my experience is lighter, gains were significant but not game-changing. Relying more on the LLM meant more iterations and adjustments. While this isn’t “the blind leading the blind,” it underscores that AI is at its best when paired with capable human leadership. The more you bring to the table, the more powerful your results with AI will be. Stay tuned: Post 5 spotlights the AI-accelerated approach we used for architectural design that helped shape our technology stack.

UX Design with AI: Does it Help or Hinder?

“LLM Tech Stack Spike Project”—that’s how Mission Control started. It was a clear sign of missing strategy. Over two weeks, dozens of focused AI sessions gave me clarity: a product vision, validated architecture, compelling value—and, most importantly, a replicable process to compress months of product discovery into days.Here’s how I transformed 50+ hours of conventional research into 15 hours of AI-driven strategy, with greater impact and less guesswork.<h3>Establish a Strategic North Star First</h3>While I knew why I was moving forward with a deep dive exploration into LLM technology, when first getting started I didn't know what I should build. I knew I wanted to truly test the limits of LLMs in real world Product Development scenarios. So I started with a North Star definition of what matters most. In the last blog post we elaborated on what these were:<ul><li>We must build on infrastructure that demonstrates strong Performance, Reliability, &amp; Scalability. </li><li>We must build on a tech stack that allows for Flexibility &amp; Extensibility. </li><li>We need to build Universal, Expected Capabilities that real world Product Development teams build in real world products. </li><li>We must use Real-World Development Workflows that would be the expectation of any team today. </li></ul>With this North Star defined, it became the basis for Product Strategy discovery / research prompts that fed into the LLMs assisting me.The decisions that I needed to make during this phase boiled down to the following:<ol><li>What would make up my technology stack? (architectural design) </li><li>What would be the best demonstration of a real-world product I should build? (product definition) </li><li>How could I best articulate WHAT I was going to build and WHY it matters? (product marketing) </li></ol><blockquote>To set out answering these questions I knew I needed to establish a strong discovery approach that also helped me maximize what I was getting out of the LLM platforms I wanted to test. Here's ultimately what worked best.</blockquote><h3>A Product Strategy Framework to Maximize Productivity with LLM Assistance.</h3>Multiple Platform "AB Testing" or "Shootouts" Relying on a single AI platform invites blind spots. Instead, we did "shootouts" or "AB tested" the major LLM platforms. Early on, we would pass the same prompts and evaluate the quality of response/returns. These were the platforms we AB tested:<ul><li>Claude Sonnet 4 </li><li>ChatGPT (o3/o4) </li><li>Perplexity Pro </li><li>Microsoft Copilot </li><li>ChatPRD (Specialized)</li></ul>I got into a good rhythm of spending time up front defining good, solid prompts, and then I'd pass the same prompt into multiple platforms / models and compare responses.Breadth--&gt;Depth Narrowing ApproachMy approach for discovery and research typically started with a breadth-first prompt. As I evaluated responses from the LLMs I would choose the best 2-3 platforms and narrow my prompting focus, moving to a more depth first prompting strategy. I would iterate on these prompts and also narrow the field of AI platforms as I went until I ultimately was able to make a final decision.It wasn't about asking more questions—it was about asking the right ones, at the right times.Breadth-first: Comprehensive prompt to all platforms to scope the landscape and identify potential leading options, based on requirements, goals, or constraints.Depth-focused: Drilled deep on the leading AI platform for technical details, use cases, pros and cons, etc on the leading options.Decision-point: Used final prompts to surface risks, confirm mitigation strategies, and ensure robust decisions.The Final Decision: Human VerifiedUltimately I had narrowed down to one LLM with a set of recommendations. From here it was important to validate and make the final decision myself. This typically involved my own deep scan of the offering or approach (whatever the decision was) and confirming it felt right. It was in these moments I realized my experience/expertise (and gut) should be the ultimate decider.<h3>LLM Productivity Gains Across Three Major Decision Areas </h3>Leveraging the research-&gt;decision framework described above, I was able to put a comprehensive Product Strategy together that included major decisions across the following areas.1. Technical Foundation Decisions Goal: Choose a technology stack and architectural approach that would be Performant, Reliable, and Scalable and give me the greatest Flexibility, and Extensibility.Time: 10 hours (vs. 20-50 hours without LLM assistance)<ul><li>Compared stacks and architectures—all platforms, framed to my constraints </li><li>Evaluated integration and risks</li><li>AI excelled at trade-off analysis, delivering actionable, context-specific recommendations (e.g., cross-platform, real-time, and scalability considerations) </li></ul>Result: Significant: 2-5x faster research, deeper analysis, catered to my North Star goals.2. Product DefinitionGoal: Determine the best Product Definition (vision, features, phased plan) that aligns with Universal, Expected Capabilities that modern Product Development teams (or technical founders) would expect to build.Time: 4 hours (vs. 6-10 hours without LLM assistance)<ul><li>Real-time market research via Perplexity Pro </li><li>Product requirements and feature prioritization with ChatPRD </li><li>Cross-AI value proposition testing </li><li>Persona validation through iterative prompts </li></ul>Result: Solid: 1.5-2x efficiency gain, stronger strategic clarity3. Value Prop / PositioningGoal: Articulate the value and positioning effectively as a product marketer would develop.Time: 2 hours (vs. 3-5 hrs without LLM assistance)<ul><li>Value proposition and elevator pitch brainstorming/ideation </li><li>Hypothetical marketing messaging to target audiences explorations </li><li>Audience segmentation aligned for each stakeholder group </li></ul>Result: Solid: 1.5-2x efficiency gain, broader creative messaging to select from<h3>What Resulted and What This Means for You </h3>By leveraging systematic, AI-powered research and decision-making, we delivered the foundation for a B2B SaaS application focused on team collaboration and communication—Mission Control. The tech stack we arrived at includes GCP/Firebase, Flutter, Hasura, and Directus CMS, all validated for scalability and efficiency.Here’s what this means for you: AI doesn’t replace your expertise—it amplifies it. With AI-driven strategic research and trade-off analysis, you gain clarity, speed, and precision that traditional methods can’t match. The result? Faster execution, better decisions, and streamlined validation across platforms.To achieve these productivity gains, adopt the framework outlined here. Integrating AI into your discovery and early decision-making processes can unlock significant advantages for your team—without sacrificing rigor or quality.&nbsp; Next up: Post 4 will dive into our experiences leveraging AI to develop Product Designs.

Weeks 1-2: How AI Accelerated Product Strategy

If you’ve spent any time browsing AI development content, you’ve probably seen the flashy headlines: “I built an app in 30 minutes!” The novelty is hard to ignore. But peel back the layers, and you’ll often find a different story beneath the surface—authentication systems held together with duct tape, zero data persistence, nonexistent error handling, and security practices that wouldn’t pass even the most forgiving audits.At Modern Product, we took a different path with Mission Control. Instead of shooting for quick, impressive demos, we designed the project with one thing in mind—production-grade readiness. The goal wasn’t to prove that an app could simply be built using generative AI; it was to prove whether it could meet the rigorous standards required of real-world, scalable software. Surprisingly, these constraints didn’t act as roadblocks—they became accelerators that guided smarter development and more efficient AI usage.Let’s start with my background and experience because, ultimately, it dictated how far I was able to most effectively leverage (productivity multiplier) versus blind trust (meager gain) in the LLMs.<h3>Who is Modern Product?</h3>I’ve spent 24 years in the tech industry, starting out as a Software Engineer before transitioning into management roles and eventually finding my home in Product Management. For over two decades, I’ve focused on Product Management and Product Leadership, working closely with Engineering and Design teams across the entire product development lifecycle. After spending 7+ years in senior and executive management positions, I founded Modern Product Consulting six years ago to share my expertise at scale.Here’s how my background connects to the “Mission Control Challenge.” As a Technology Leader, I’m comfortable discussing complex engineering topics just as easily as I can tackle strategic business conversations with executives. First and foremost, I consider myself an expert in Product Management and Product Strategy. I’d call myself an intermediate full-stack Software Engineer and a solid QA practitioner, but I have no real skills in infrastructure or DevOps. I’ve worked alongside some incredible UX Design professionals, but I’m not a designer myself.&nbsp;When I took on this Mission Control challenge, I knew there would be areas where I could lean on my own expertise and others where I’d need significant support from LLMs to build a production-ready application from scratch. That’s the key—our insights here are driven by where my background allowed me to amplify productivity and where I had to allow and trust the LLM to step in to fill the gaps.<h3>So...What is Mission Control Exactly?</h3><figure id="figure-ch71l2brssmeaw24x5"><img id="ch71l2brssmeaw24x5" src="https://cdn.durable.co/blocks/19IRM6Nz9ajbOwp8j44BJfjSelTetqqJXyZfVncNVjkxALS2Qg8YmllAX9efmdSN.png" alt="" title="" class="object-cover bg-gray-100 rounded-sm md:rounded-md lg:rounded-lg" style="aspect-ratio: 3/2; object-position: 50% 0%;" data-corners="default" data-aspect="3:2" data-caption="" data-positionx="50%" data-positiony="0%" data-fullwidth="false" data-mce-src="https://cdn.durable.co/blocks/19IRM6Nz9ajbOwp8j44BJfjSelTetqqJXyZfVncNVjkxALS2Qg8YmllAX9efmdSN.png" data-mce-style="aspect-ratio: 3/2; object-position: 50% 0%;"></figure><blockquote>Imagine your entire team’s activity—notes, events, requests—flowing through a single, beautifully organized workspace. Instantly see what matters: smart tags, AI-surfaced insights, a real-time pulse on who’s doing what, and weekly digests that zero in on actions, decisions, and blockers. This isn’t just collaboration; it’s your team running at its sharpest, with the chaos streamlined and the signals amplified.&nbsp;</blockquote>(ChatPRD-generated Product Marketing messaging)<figure id="figure-6sqnw6bkwvjmeaw3waz"><img id="6sqnw6bkwvjmeaw3waz" src="https://cdn.durable.co/blocks/2bUrxS8QEp8tzwKvt2d3kz73iibf3KnQtbVY2VjCNwXqh68RdDlvw8lG2ZauWp50.png" alt="mission control notes" title="mission control notes" class="object-cover bg-gray-100 rounded-sm md:rounded-md lg:rounded-lg" style="aspect-ratio: 3/2; object-position: 50% 0%;" data-corners="default" data-aspect="3:2" data-caption="Mission Control Notes" data-positionx="50%" data-positiony="0%" data-fullwidth="false" data-mce-src="https://cdn.durable.co/blocks/2bUrxS8QEp8tzwKvt2d3kz73iibf3KnQtbVY2VjCNwXqh68RdDlvw8lG2ZauWp50.png" data-mce-style="aspect-ratio: 3/2; object-position: 50% 0%;"><figcaption class="body-normal text-gray-500 mt-2">Mission Control Notes</figcaption></figure>Mission Control is a B2B SaaS application that supports multiple customers with multiple users, all in hypothetical team-collaboration scenarios. It showcases collaboration features like shared notes, requests management, events management, team documentation, and public blogs managed across a team. We wanted to demonstrate how AI/LLM tech can boost the Product Experience as well, so in the Mission Control app, we built AI-generated "Weekly Summary," personalized "Action Items" or "Notifications" to individual users, and automated AI-based tagging across all content that feeds into search and team-based productivity stats/reporting.&nbsp;<figure id="figure-a67yavoogfmeaw5ztc"><img id="a67yavoogfmeaw5ztc" src="https://cdn.durable.co/blocks/18swM6ZMjmqB5rwvUbfcTaocc6GbzwHjW7tiTweuo9qR6CkF996tDO3JuXqdj1mR.png" alt="mission control blog" title="mission control blog" class="object-cover bg-gray-100 rounded-sm md:rounded-md lg:rounded-lg" style="aspect-ratio: 3/2; object-position: 50% 0%;" data-corners="default" data-aspect="3:2" data-caption="Mission Control Blog" data-positionX="50%" data-positionY="0%" data-fullWidth="false"/><figcaption class="body-normal text-gray-500 mt-2">Mission Control Blog</figcaption></figure>(NOTE: Mission Control is a production-ready application that we aren't going to take public. It’s solely built to showcase the transformative possibilities of AI and large language models (LLMs) in modern product workflows, and we hope to serve as a foundation tech stack for founders who wish to springboard into their new venture product development. We wanted to highlight how AI/LLMs can boost productivity and speed up time-to-market, but do so in a real-world production environment.)<h3>The Production Realities That Guided Mission Control </h3>When building a business-critical application, flashy UI elements will never outweigh the need for robust functionality and reliability. Mission Control wasn’t going to be another toy project. It had to solve the same challenges your team faces when scaling up from idea to execution. Here are the key production requirements that defined the project.Performance, Reliability, &amp; Scalability Performance, reliability, and scalability are crucial for building successful products. Prioritizing these early ensures systems run efficiently, handle growth, and stay dependable over time—giving companies a strong foundation for sustainable success in a competitive market.Flexibility &amp; Extensibility A flexible tech stack is key for any forward-thinking organization. As needs and strategies evolve, choosing adaptable tools ensures long-term success. By selecting a tech stack that supports growth and change, businesses can innovate and stay competitive.Universal, Expected Capabilities Foundational features are key to aligning a product with its strategy and meeting user expectations. In B2B SaaS, app architecture must support multiple customers, ensure data security, and include essentials like user accounts, login flows, and role-based access. These "table stakes" must be perfected before adding advanced features to deliver a reliable product.Real-World Development Workflows Efficient development pipelines are key for scalable, collaborative production teams. Source control tracks changes and streamlines teamwork, while build scripts automate deployments. Clear configuration files maintain consistency, and thorough README documentation helps team members navigate projects easily. Containerization ensures reliable, environment-specific workflows, reducing “it works on my machine” issues. Together, these practices enable sustainable software development.These aren’t just hypotheticals—they’re the foundations for how Mission Control was designed, built, and deployed.<h3>How Constraints Accelerated Development </h3>1. Rigorous Planning Led to Clearer Specifications and Better Outputs&nbsp; Modern product development relies on smart planning and clear specifications while executing as lean and efficiently as possible. Applying this rigor to AI-assisted workflows paid off significantly. Well-defined requirements gave LLMs clear direction, avoiding vague or subpar outputs. For instance, specifying detailed user stories and edge cases allowed the models to generate more focused, high-quality code and content.&nbsp;Measurable Benefit: Reduced rework significantly (ballpark 5–10x) compared to using generic prompts, speeding up delivery timelines.2. Leveraging a Modern Tech Stack Boosted LLM Effectiveness&nbsp; By using well-supported, modern tech stack components, the LLMs had access to a larger and more reliable base of training data and expertise. Familiar stacks like Flutter, GCP/Firebase, Docker, and well-supported languages like Dart and Python were within the models' knowledge base, which enhanced the quality of their solutions. This alignment reduced gaps in generated outputs and improved compatibility with real-world standards.&nbsp;Measurable Benefit: Increased engineering time exponentially (10–50x) and decreased troubleshooting time (at least 5–20x), with AI generating production-ready suggestions more often.3. "Shootout" Testing Optimized LLM Selection for Each Task&nbsp; Just like we would in real-world product development when evaluating new platforms or technologies, we took a "shootout" approach to testing various platforms/LLMs like Claude Sonnet, ChatGPT 4, Perplexity Pro, and CoPilot against each other for various tasks. By A/B testing their performance, we identified which model was best suited for specific needs, such as content generation, coding/debugging, DevOps, or architecture planning.&nbsp;Measurable Benefit: Relative to when we started, our workflow efficiency improved 2x, as we matched tasks to the best-performing model, minimizing wasted effort and maximizing output quality.<h3>Actionable Takeaways for Tech Leaders &amp; Teams&nbsp; </h3>Use your Team's Expertise to Guide AI&nbsp;With 200 hours of deep exploration, one thing is clear: having experts "behind the wheel" is critical to truly unlocking LLM potential. Treat LLMs like a "gifted intern"—they need monitoring and guidance, but when pointed in the right direction, they can produce incredible results and exponentially boost productivity in specific areas.&nbsp;AI Won’t Replace Expertise, But It’s Still Valuable&nbsp;AI isn’t ready to replace skilled professionals, but even in areas where expertise is lacking, you can still achieve marginal productivity gains of 25–100%. While it’s a worthwhile tool to augment your team, it’s not a reason to eliminate your "doers." Skilled professionals remain essential for critical thinking, contextual decisions, and creative problem-solving.&nbsp;Strategic Processes Drive Better Results&nbsp;The effectiveness of LLMs for product development improves significantly with strong planning, clear requirements, and well-defined specifications. A strategic approach to prompting, learning which LLMs excel at specific tasks, and iterating along the way will maximize the benefits of integrating AI into your workflows.&nbsp;By leveraging AI strategically and complementing it with human expertise, you can unlock precision and speed in your product development process—without compromising quality.&nbsp; NEXT POST: We'll walk you through the early Product Strategy/Product Management work we did and what we learned in working with various AI platforms/LLMs.

Reality versus "Demo Magic"

What if you could build a production-ready app—on your own—in just ten weeks? Not a prototype, but a robust application with the kind of professional polish that could scale and adapt to real-world business needs. That’s the challenge I set for myself with the Mission Control project. It wasn’t just to see if it was possible, but to prove how the capabilities of Large Language Models (LLMs) can transform the way we approach product development.This series is for those of you leading technology-driven ventures—founders, CTOs, VPs of Product, and engineering leaders—who aren’t just watching the AI revolution but are eager to harness it to accelerate your own goals. I’m here to share not a theoretical deep-dive into AI’s potential, but a tangible, real-world blueprint for how you and your teams can push boundaries and achieve extraordinary results with LLMs.<h3>The Origins of the Mission Control Challenge&nbsp; </h3>For years, I’ve helped tech teams and product-focused organizations refine their strategies, overcome bottlenecks, and optimize delivery processes. Modern Product, my consultancy, was built on the foundation of guiding organizations through complex product challenges. But in early 2025, I noticed that AI-driven tools—and specifically LLMs—were beginning to make extraordinary claims. "Write all your code with AI." "Revolutionize the speed of development overnight." And yet for all the hype, there was a persistent question from leaders I worked with:<blockquote>Sure, we’ve seen the demos, but how do we actually implement AI in our real development workflows without compromising standards?</blockquote>The Mission Control challenge emerged as my answer to that question. If tools like GPT-4 or Claude can boost team productivity, why couldn’t they also help a solo product expert like me, with strong technical and design acumen, do the work of an entire startup team? My constraints were clear from the beginning:<ul><li>10 weeks. A tight but realistic timeline to prove the capabilities of LLMs under real-world pressures.&nbsp; </li><li>200 hours. That’s what I limited my dedicated time to—about 20 hours a week between other responsibilities.&nbsp; </li><li>A production-ready app. This wasn’t an experiment nor a throwaway prototype; the output needed to demonstrate scalable, secure, and usable industry-grade features.</li></ul>&nbsp;It was a bold goal. But it also seemed like the perfect way to test the limits and learn the lessons an entire industry is earnestly seeking.<figure id="figure-tp14paqxzzmdm9se9k"><img id="tp14paqxzzmdm9se9k" src="https://cdn.durable.co/blocks/19IRM6Nz9ajbOwp8j44BJfjSelTetqqJXyZfVncNVjkxALS2Qg8YmllAX9efmdSN.png" alt="mission-control-web-app" title="mission-control-web-app" class="object-cover bg-gray-100 rounded-sm md:rounded-md lg:rounded-lg" style="aspect-ratio: 16/9; object-position: 50% 0%;" data-corners="default" data-aspect="16:9" data-caption="Mission Control" data-positionX="50%" data-positionY="0%" data-fullWidth="false"/><figcaption class="body-normal text-gray-500 mt-2">Mission Control</figcaption></figure><h3>Why LLMs Are a Game-Changer for Product Development </h3>Using artificial intelligence for small, task-specific operations has been possible for years. But with LLMs, we’re talking about something altogether different. Tools like OpenAI’s GPT or Anthropic’s Claude have unlocked capabilities that feel almost human-like in their ability to interpret complex prompts, tackle deep workflows, and adapt on the fly.And here’s the game-changer—they’re not just tools; they function like "intelligent interns." They’re brilliant one day, baffling the next, and always in need of guidance. LLMs are never wholly autonomous, but they become extraordinary amplifiers when paired with the right oversight. Here’s why they matter:<ul><li>Speed: LLMs can accelerate tedious workflows, transforming tasks that take hours into quick iterations. </li><li>Ideation Partner: From brainstorming features to flagging bugs, LLMs can act as a collaborator to challenge assumptions and refine ideas. </li><li>Scalability: For lean teams—or even solo developers like me—they lower the hurdle to achieving outcomes usually requiring entire teams. </li></ul>Yet, as every technology leader knows, speed means nothing if it comes at the cost of quality, security, or long-term maintainability. This dual promise—faster development and industry-grade results—was the hill that Mission Control set out to climb.<h3>Why This Series Matters to You&nbsp; </h3>With over 20+ completed features, countless learnings, and hard-won insights about what LLMs can (and can’t) do, Mission Control represents more than just an experiment. It’s a scalable process framework—something product leaders and tech teams can dissect, adapt, and apply to generate net-positive outcomes.Here’s what you can expect from reading this series:<ul><li>Unfiltered Lessons Learned: I won’t gloss over the moments where things fell apart. When debugging a legacy issue took 3 hours instead of 30 minutes, or when the LLM’s solution led to circular dead-ends, you’ll hear about it. </li><li>Practical Strategies: From prompt engineering to designing LLM-assisted workflows, I’ll break down the approaches that worked (and why they did). </li><li>Expert-level Solutions: Whether it’s testing and QA using LLMs or balancing AI-generated output with human oversight, the posts will focus on achieving high-quality, high-value outcomes. </li><li>Future-Proof Takeaways: You’ll leave with frameworks and tools to help your team make smarter decisions on how to incorporate AI into your workflows without lowering your development standards. </li></ul>Whether you’re trying to validate an idea in record time, expand your team’s bandwidth, or build scalable solutions from Day 1, this series is designed to inspire and guide you.<h3>What’s Next? </h3>Over the coming weeks, I'll take you through each phase of the Mission Control challenge. Topics include:<ul><li>How I explored the LLM landscape to select the right tools and frameworks.</li><li>The unexpected ways prompt engineering influenced product design. </li><li>Successes and failures in engineering, debugging, and QA. </li><li>Reflections on scope management and balancing AI strengths and weaknesses. </li></ul>By the end, you'll have actionable strategies and a clear understanding of how LLMs can become a critical lever for innovation while avoiding the pitfalls that trip up so many early adopters.The Mission Control app is proof of what’s possible when cutting-edge technology meets disciplined product expertise. And if you’ve been searching for a roadmap to harness AI in your own workflows, you're in the right place.&nbsp;Stay tuned for insights that will challenge the hype, inspire innovation, and help you scale with confidence.<h3>Call to Action </h3>Curious about how your team could amplify productivity and maintain quality with LLMs? Feel free to reach out. Whether it’s a consultation or a conversation, I’d love to help unlock what’s possible for your next product.&nbsp;

Harnessing LLMs for Product Development: The Mission Control Challenge

<blockquote>...this celebration is not just about looking back at what we've accomplished; it's also an opportunity to reflect on the lessons learned along the way.</blockquote>This month marks a significant milestone for Modern Product as we celebrate our five-year anniversary. We've had the privilege of working alongside incredible clients on a diverse array of innovative and challenging projects. We've been extremely fortunate to remain at full capacity the entire time, with a 100% engagement renewal rate across 15 distinct clients!That said, this celebration is not just about looking back at what we've accomplished; it's also an opportunity to reflect on the lessons learned along the way. Today, we'd like to share some of the insights we've gained in the last five years. Through three key lessons, we hope to provide fellow founders with valuable guidance and encouragement that can help them on their own journeys.<h3>Lesson 1: Embrace Ambiguity and Stretch Yourself </h3>When Modern Product was just getting off the ground, we quickly realized that no two projects would be the same and that there would be aspects of every engagement that would take us out of our comfort zones. If we were going to see success as a consulting practice, we recognized the need to embrace that ambiguity and look forward to those opportunities to stretch ourselves. By leveraging our expertise as a springboard into new challenges and new opportunities to learn, we were encouraged to seek out diverse engagements with clients across different industries. Now, we always look forward to kicking off new engagements, for the opportunity to face the ambiguity and learn along the way.&nbsp;So our advice to new founders is to make full use of your expertise and experiences, not as a constraining box that limits your focus, but as a springboard into the ambiguity that's out there.<h3> Lesson 2: Commit to Business Development 365 Days a Year </h3>One of the earliest commitments we made for Modern Product was to continually invest time in business development. When Modern Product was just getting off the ground, we had no idea how client engagements would pan out, nor how often new client opportunities would arise. We received the advice that all consultant founders get, that we would need to be prepared for the "feasts and famines" of client work. One of the best decisions we made early on was to commit 5 hours per week to business development activities as a preventative measure against the feast/famine swings. We've maintained this practice over the years, and it has proven to be worthwhile several times over. This commitment has played a pivotal role in our sustained growth and success over the last five years.Our advice: never turn down an introduction, invest time in all forms of networking, even the most casual coffee chats, and apply success metrics to these activities to hold yourself accountable.<h3>Lesson 3: Enjoy the Surprises Throughout the Journey </h3>One of the most rewarding aspects of being a founder is experiencing the unexpected surprises that come along the way. The entrepreneurial journey is filled with twists and turns, and it's important to savor these moments and appreciate the growth they bring. Over the five years of Modern Product's existence, we've had several surprises pop up that, in hindsight, have been pretty cool.&nbsp;<ul><li>We landed a client engagement with a dream brand.&nbsp; </li><li>A random coffee chat about Modern Product's business model unexpectedly led to a client engagement four years later. </li><li>We've reconnected with old colleagues and classmates from 10, 15, 20 years ago.</li></ul>Looking back, the surprises have been a delight to reflect on, making the entrepreneurial journey of being a founder that much more rewarding and rich.&nbsp;So our recommendation to founders just getting started: enjoy the journey's unexpected turns and dedicate time to looking back and reflecting on the good, the bad, and the ugly along the way. Treat everything as a learning opportunity.<h3>Conclusion </h3>Being a founder is as challenging as it is rewarding. The past five years have taught us that while the road may be unpredictable, it's essential to enjoy the journey and focus on the learning opportunities it presents. Each client engagement has contributed to our growth and success. Hopefully this post provides encouragement and inspiration to those of you diving into the founder consulting journey!

Blog

How AI Boosted the Architectural Design Process for Missi...

UX Design with AI: Does it Help or Hinder?

Weeks 1-2: How AI Accelerated Product Strategy

Reality versus "Demo Magic"

Harnessing LLMs for Product Development: The Mission Cont...

On Celebrating Five Years...