I've built across the stack — turbine engines, embedded firmware, cloud platforms, data systems, and lately a lot of AI. The layer changes; the instinct doesn't: learn it well enough to build with it, then make it better.

Most of what I build is private — client and company code. The repos here are the showable tip: small tools, but they run, and they're the kind of thing I build everywhere.

A few worth a look

• slack-aws-cost-guardian — Serverless AWS cost monitoring with AI anomaly detection, pushed to Slack with a bot you can @mention. CDK-deployed, ~$5/month. The cost discipline I want on every project, automated.
• vector-database-loader — The plumbing behind good RAG: populate a vector database from messy real-world sources, with the chunking and filtering that decide whether retrieval actually works.
• smart-artifact-parser — Turns raw source artifacts into clean, structured context an AI can use — the front of the RAG pipeline, where retrieval quality is usually won or lost.

Elsewhere

• The work, the writing, the rest → danjamkuhn.com
• I write about AI and engineering leadership on Medium