Quick hits on product craft, AI, and tech curiosity—designed to be useful, not noisy 🤝🤖🧠🧪
Building an ExperimentOps Lite tool to demonstrate SQL analytics, workflow design, and practical AI integration—from intake through decision, with DB-enforced integrity and cached AI readouts.
How I built Cryptopunk, an open-source blockchain data platform that combines real-time detection with explainable risk scoring to protect users from Solana rug pulls.
Codex Safe is a protocol for designing safe, aligned LLM-in-IDE code assistants that touch real production code and infrastructure.
Hellfire Mode is a product principle for AI systems that prioritizes truth, uncertainty, and risk over comfort.
A practical control layer for AI behavior—defining authority, reality, contracts, oversight, and epistemic honesty so assistants and agents stay aligned, observable, and interruptible.
