AI & LLM Systems .

Six specialized agents grade in parallel under Temporal, and a confidence gate sends only the shaky grades to a human.

The grading platform I designed after 400+ Java and Python submissions landed on my desk as a TA. Six specialized agents grade in parallel under Temporal orchestration, a confidence gate routes only the shaky grades to a human, and it runs local-first, so student work never leaves the machine.

An LLM gateway that matches prompts by meaning, so 78 percent of requests never reach the model.

A gateway that remembers what your model already said. It matches prompts by meaning, not exact text, so 78 percent of requests skip the model entirely and the typical answer comes back in 44 milliseconds. The memory runs on Proxima.

Guards what agents can do: a default-deny policy on every MCP tool call, and tool output checked before the model reads it.

A Rust gateway for the Model Context Protocol. It sits between an agent and its MCP servers, refuses any tool call its policy did not permit, reads tool output before the model does to redact credentials and fence off injected instructions, and writes an audit record for every decision. Default deny, 224 tests, CI on Linux, Windows and macOS.

A model learns a back-office task once; the recorded capability then replays with no model in the loop.

Computer-use automation for back-office web apps that have no API. A model works out how to finish a task in a legacy app once, and the run is recorded as a typed capability that replays deterministically with no model in the loop, tells a business outcome apart from a broken app, and hands the live browser to a person when it meets something it does not know.

Answers free-text requests like "Arrival but funnier" with a two-tower retriever and a neural ranker.

A two-stage hybrid movie recommender, built the way production systems are. A two-tower neural model retrieves candidates through Proxima, my own C++ vector search engine; a neural ranker reorders them. It answers free-text requests like 'a slow-burn sci-fi like Arrival but funnier' and includes movies in theaters this week.

A study assistant grounded in real sources: every answer carries citations, and it says so when it does not know.

A study assistant that refuses to make things up. Every answer is pulled from real source material and arrives with citations, and it stays quiet when it genuinely does not know.

An AI patient assistant in text and voice, taken from an empty folder to shipped in under a week.

An AI patient assistant that talks back, in text and in voice. I took it from an empty folder to a shipped product in under a week, working entirely AI native.

A separate agent on Workers AI writes a brand new mini game every day.

A real-time multiplayer PWA I built for my long-distance girlfriend and me. Surprise games drop onto both phones at random times when we're both awake, we play live for a minute, and a separate AI agent writes a new game every day. One Durable Object per room, Web Push written from scratch, and it runs for $0 on Cloudflare's free plan.