Daniel Schmidt

great point, but what mazelab is actually doing is a bit different i think to explain how it works, every prompt i send to a coding agent gets logged automatically via MazeLab's MCP. each session records the goal, what actually happened, and the outcome. that data is published publicly so anyone can read the real process of building with AI — not just the finished result the focus is making it publicly legible

this is essentially what persistent memory MCP servers do automatically - each session the AI logs what was built, decisions made, the why. next session it recalls all that context without you having to write the retro yourself. you're solving the same problem, just manually right now

I'm building in public, and now my AI writes the retrospective. Every vibe coding session: what I wanted to build, what actually shipped, and what I should do differently next time — auto-updated, all public. My hope is that real, unfiltered prompt examples are more useful than polished tutorials. And that more people start sharing the same. mazelab.ai

@WebcraftGallery lol, that's fair

@danielfschmidt i cant say i understand what that's about

The new “build in public” is publishing your AI prompts + what actually happened. I’m trying it here: mazelab.ai

Ask your coding agent this: "Imagine you are a human engineer who has building this app. What would be a surreal dream you would have that speaks to the substance of what you are building, and the fears/anxieties/wishes you have related to it?"

Having a personal blog feels low leverage now. So I created this personal changelog: mazelab.ai Through my changelog's MCP server, the agents building my various experiments connect to it and log their goals and learnings automatically. It's now the most up-to-date representation of what I'm actually doing and thinking about.

You’re shipping features, running experiments, and tracking dashboards full of data, but still wondering: Is any of this actually moving the needle? The data is sobering: Only 10-30% of shipped features create measurable business impact. That’s the messy middle: the gap between everyday work and real business impact. For years, only the biggest companies could solve it by throwing massive resources at the problem. Now that’s changing. Our new interactive manifesto from @danielfschmidt explores how value architecture, powered by data-connected metric trees, makes it possible for any team to connect their work directly to business outcomes and finally see where value is really created. Unlock the web experience 👇 mixpanel.com/content/messy-…

We founded @DoubleLoopApp to help teams escape the feature factory & drive impact. Now we’re excited to share that we're joining @Mixpanel — a team that shares our vision. 🚀 👉Read more about our journey and why we’re joining forces with Mixpanel: mixpanel.com/blog/doubleloo…

@markhalperin @ezraklein You are the two lenses I rely on the most to understand politics. Would love to see the two of you make content together.

We're developing AI at @DoubleLoopApp to help product teams predict and quantify the impact of their bets before committing resources. Here's a 5-min demo. Would love your thoughts. loom.com/share/735d03ce…

𝗧𝗵𝗲 𝗔𝗰𝘁 𝗼𝗳 𝗖𝗿𝗲𝗮𝘁𝗶𝗼𝗻 Connecting two seemingly disparate things can sometimes yield profound insights. Listen to @danielfschmidt talk about 'The Act of Creation' and learn why connecting your short-term inputs to your long-term outcomes is a great place to start for Tech. executives. The field of management is inherently multidisciplinary, and integrating ideas & insights from other fields has the potential to profoundly transform it. In my 23 years of corporate experience, I committed so many mistakes and category errors just by embracing the prevalent management practices without thinking. But, this is not surprising! It took us over 5000 years to apply the concept of ‘wheels’ to luggage. This underscores our human tendency to be ‘domain dependent’. We often struggle to transfer insights and solutions across different disciplines, even if the application seems obvious in retrospect. Join the Cyb3rSyn Community to explore these ideas and listen to the full podcast episode with Daniel - link in the post below 👇🏾

𝐓𝐡𝐞 𝐌𝐞𝐬𝐬𝐲 𝐌𝐢𝐝𝐝𝐥𝐞 Do you experience what @danielfschmidt calls the "messy middle," where many people doing the essential work — engineering, design, marketing — struggle to understand how their work fits into the bigger picture of the business? Does your work often feel driven by the whims or intuitions of those above you? What can Tech. Executives do about it? Listen to the inaugural Cyb3rSyn Labs podcast, exclusive to the members of the Cyb3rSyn Community - link below 👇🏾

@danielfschmidt Link: cyb3rsyn.com/p/dealing-with…

This week’s Cyb3rSyn Newsletter summarizes the key highlights of the inaugural Cyb3rSyn Labs Podcast episode - my conversation with @danielfschmidt, the Founder of DoubleLoop. I wanted the first episode of the Cyb3rSyn Labs Podcast to be at the intersection of the ideas that we discuss in the Cyb3rSyn Community (Systems Thinking, Complexity, Cybernetics) and Entrepreneurship - I couldn’t think of anyone better than Daniel! Daniel’s startup, DoubleLoop, helps product development teams to build products in ways that embrace the insights from these fields to avoid the typical pitfalls of the mainstream. Topics Discussed: ✅ Focus on Input Variables ✅ Messy Middle & Aligned Autonomy ✅ Outcome-Oriented Product Development and much more... #entrepreneurship #complexity #systemsthinking #leadership #productmanagement

@finn_meeks @finn_meeks love it. Simulations have become the frontier of what we're building at @DoubleLoopApp with our new "Bet Simulator." I recent wrote this on the topic: linkedin.com/posts/danielfs…

Another axis of curiosity. I'm hoping to work with 1-2 companies in this space in the upcoming fellowship. What will simulation-driven product development look like? Simulations help us understand and predict the world in various domains. Game engines and physics models are improving rapidly and could transform how we understand, predict, and interact with the world. - How close are game engines to achieving physics-realistic simulations suitable for high-quality data generation? - In what new domains could these advanced simulations be applied (e.g., personalized medicine, urban planning, climate modeling, etc.)? - How will simulations impact product development cycles? - Will AI's reasoning capabilities unlock new frontiers in science? @gopalkraman also hosted an excellent roundtable on ML and simulation a few days ago -- we have many folks in our community also eager to meet you.

@AnnieDuke we're building a new "Bet Impact Simulator" for product management teams at @DoubleLoopApp . I've used your concept of "Resulting" to frame the problem we solve. See here: linkedin.com/feed/update/ur… Let me know if this is something that you're compelled to dig more into.

8/ With AI reducing the friction of acquiring domain knowledge, the gap between wanting a system to work better and building it is shrinking. Passion and vision are still critical, but the barriers for inside-out neophytes to execute are lower—and falling every day.

7/ I’m biased, but I believe the advantage increasingly goes to inside-out neophytes thanks to LLMs . Conversing with AI has helped me advance up the data science learning curve faster than I ever could have before. Without it, DoubleLoop might not exist today.

1/ Do you agree with the idea that neophytes are the best people to found companies? The argument is that individuals with less domain experience are uniquely positioned to revolutionize an industry because they’re not constrained by conventional thinking. 🧵