I build reliability systems and AI agent tooling.

Over the last decade I’ve worked on production infrastructure at scale: SLO platforms for 1,000+ microservices, topology-aware anomaly detection, automated RCA, and JVM/platform migrations across large fleets. Lately I’ve been focused on agent orchestration, developer tooling, and small systems that are useful enough to ship.

skillpm Universal package manager for AI agent skills. Install, sync, and secure skills across Claude Code, Codex, Gemini CLI, Copilot, Cursor, and other runtimes.

lindy-orchestrator Agent orchestration framework for DAG scheduling, QA gates, worktree isolation, and git-native coordination.

claude-insight Rust-based tooling for inspecting Claude Code sessions, traces, and replay evidence.

moneywiz-ledger A practical personal-finance skill: turn natural-language expense entries into MoneyWiz deep links with local config and category inference.

I try to keep the public proof close to the code.

• NousResearch/hermes-agent — reported and helped close model-routing/cache compatibility bugs in the auxiliary client path: issue #5803, issue #5809
• microsoft/VibeVoice — fixed the streaming TTS SDPA regression on MPS/CPU: issue #312, PR #303
• nodejs/undici — fixed cache deduplication key collision behavior: issue #5012, PR #5013
• whiteher0n/option_picker — 17 merged PRs across intelligence, events, risk, and streaming analysis work: merged PR search
• Other accepted reports include mlx-lm

• AI agent tooling
• reliability and observability
• developer infrastructure
• Python, Go, Rust, TypeScript, Java