software engineer · ai agent systems
i build reliable systems for AI agents.
harness juggler. rl environments, orchestration, agent-safety, and the eval harnesses that keep fleets of agents honest.
// about
about
i build the systems ai agents run on: rl environments, eval harnesses, multi-agent orchestration, and the reward design and agent-safety that keep them honest. reward that resists goodhart; safety that's architectural, not just prompt-level.
before that, three years shipping web and smart contracts. i like systems that are boring enough to trust: explicit over implicit, signal without noise, the minimum code that solves today's problem.
// skills
skills
languages
ai / agents / eval
tools
backend / systems
infra
// work
selected work
claudima ↗
production multi-agent platform in rust: json-schema-typed tools, a sandboxed subagent-spawning primitive, and a two-process agent-safety model that structurally blocks prompt injection from reaching code execution.
foundry ↗
solo full-stack parametric jewelry cad studio. next.js + fastapi + cadquery, with a castability engine.
open source ↗
tooling and harness extensions to open-source agentic-evaluation frameworks like harbor and terminal-bench.
// activity
activity
// writing
writing
// work with me
work with me
pick what fits, then book below.
intro
15m · freequick fit-check. is this worth both our time?
agent reliability audit
paidi pressure-test your agent stack for reliability, prompt-injection, reward-hacking, and eval gaps, then hand you a prioritized report.
harness / eval build
projecta custom eval harness, benchmark, or rl environment for your agents or models.
working session
60m · paidbring a live problem: flaky multi-agent, reward getting gamed, sandbox design. we fix it together.
hiring
30m · freeyou're hiring for agent-infra or eval and think i fit.
embed not loading? open the booking page ↗
// contact
contact
the fastest ways to reach me: