All Writings & Insights
A catalog of verified tactical checklists, strategic research audits, and structural frameworks exploring advanced AI ecosystems and data science.
I Deleted Half My Agent's Instructions, and Here's What Happened
729 blind trials on CLAUDE.md and skill files. The same document scored 87% and 13% depending on nothing but the filesystem underneath it.
What Form Should Knowledge Take?
The evidence from AI frontier settling on what curated knowledge should look like as a system. And a new place for RAG and retrieval turns out to serve it.
How I Think About Knowledge in Finance
Finance spent two decades improving its data systems. It did far less to preserve the reasoning behind decisions. AI has lowered the cost of finding that reasoning. This essay explains how financial firms can manage it.
Building OptiMind: Memory Architecture for Claude Code on Mobile
Agentic memory is a harder problem than it looks. Here's what I learned building a memory-enabled agent around GitHub and Claude Code.
Let the Flywheel Spin on Top of the Translation Layer
How 6+ years of building, shipping, and 40+ AI rooms came together to uncover a truth about earning trust in AI
A Science-Based Skincare Routine
People keep asking what products I use. The answer is boring. The principles behind them aren't. A pairing of my daily routine with the dermatology and photobiology research that justifies each step.
Tracing the Minds Behind Claude Code
I spent a weekend reading two Claude Code source trees side by side, to find what changed and hasn't across 13 months of the most successful AI product.
Agents Don't Need Better Models. They Need Better Infrastructure.
I watched 300 builders hit inference limits at Mistral's worldwide hackathon. Three weeks later, GTC 2026 announced the infrastructure to fix it. Here's what agent developers need to understand about the serving layer beneath the models.
What Senior Business Leaders Actually Ask About AI
I guest-lectured to Directors and C-suite leaders at Columbia's executive AI program. Their questions, about liability, talent, and competitive advantage, reveal where leadership's mind actually is on AI right now.
How I Learn Anything with AI
The feedback loop that used to require study groups, office hours, and weeks of iteration now happens in real time. How arguing with Claude turned six weeks of Causal Inference into a repeatable system for learning anything technical.
Connecting Local to Global: Inside Mistral AI's Worldwide Hackathon
7,000+ applications, 1,000+ builders, 7 cities, $200K+ in prizes: inside Mistral AI's first worldwide hackathon, from the judge's seat at the New York edition.
The AI Superpower for PMs: Why Context Engineering Is the Skill That Changes Everything
Context engineering (not prompt engineering) is the PM superpower that separates AI-native product managers from everyone else.
How We Brought 300 Builders to Manhattanville on Valentine's Day
300 builders, 60+ teams, 8 corporate sponsors, and a 20-person organizing team. Inside the Valentine's Day hackathon at Columbia Business School's Geffen Hall.