War stories from the trenches. Real problems, real solutions, real lessons learned.
Reusable AI assets. Download and adapt to your own needs.
Curated learning paths for specific AI topics.
Longer-form analytical pieces with arguments, evidence, and conclusions.
Why This Exists.
AI capability is advancing rapidly. Our ability to apply it meaningfully is not.
The "AI experts" I encounter fall into two camps: machine learning engineers who've never run a business, and business people who use AI extensively but don't understand how it actually works under the hood.
This gap won't resolve quickly. We can't wait 20 years for today's ML graduates to acquire domain expertise. Instead, people with deep domain knowledge need to build real AI competency now.
Not "knows 10 ChatGPT tricks." Real competency means understanding what an LLM actually does. What persists across conversations and what doesn't. How tool use works. What RAG and embeddings are for. Token economics. Evaluation. The boring nuts and bolts that make AI useful rather than just impressive.
Andrej Karpathy talks about LLMs as an emerging operating system: software 3.0. That gave me language for something I'd already been feeling. The Claude Projects I build aren't just workflows, they're programs. I want to share them the way developers share code. I want the learning plans I create to be reusable assets.
Excellent, no-BS information about AI exists. It won't come to you passively. What comes passively is "Top 10 AI Tricks." Skin-deep content that doesn't advance our species.
This Lab is my attempt not to add to that noise. I'm sharing what I build: case studies of real problems (including what didn't work), reusable assets, learning paths I've followed, and occasional deeper analysis. I'm not solving the expert gap. I'm documenting what I'm actually doing, in enough detail that it might be useful to someone else.
What's Here: Case studies walk through specific problems I've solved with AI: what worked, what failed, and why. Prompts & Projects are downloadable assets you can use directly. Learning Plans are the sequenced resources I followed to skill up on specific topics. Research & Analysis contains longer-form pieces when I have something substantive to say. Current Explorations shows what I'm actively working on right now.
