Systems Architect · Research Engineer · Autonomous Systems Developer
Self-taught. Multi-repo open-source contributor. Python, graph tooling, ML infrastructure. Using AI to rapidly understand; diagnose and repair architecture
I contribute production-grade fixes and features to established open-source projects. My work spans graph databases, RL training infrastructure, and developer tooling, with a track record of getting PRs merged by real maintainers on real codebases.
focus:
- Python (strict typing, mypy, ruff)
- Graph databases & knowledge graphs (Neo4j)
- ML/RL infrastructure
- Developer tooling & automation
- Git/GitHub workflow engineering| Project | What I shipped |
|---|---|
| neo4j-labs/neocarta | Context-manager support across all connectors · Fixed unreliable embedding-write counters · Hard-coded BigQuery platform metadata |
| neo4j/neo4j-graphrag-python | Added structured output support to AnthropicLLM |
| JumpMasters/quartermaster | Fixed N+1 query pattern in order repository · Bounded request field lengths for API safety · Shaped public API error responses |
| qorexdevs/quicksave | Added diff --name-only and find --count CLI features |
| SikamikanikoBG/homelab-monitor | Built TLS certificate expiry tracking · Maintenance-window alert silencing |
| verl-project/verl (ByteDance) | Atropos RL environment integration with GRPO training (open, awaiting maintainer review) |
| HolmesGPT (CNCF SRE Agent) | 3 open PRs awaiting review — tool-search integration, turn-handling fixes, HealthCheck execution ordering |
All contributions are independently verifiable via merged PR history — no inflated claims.
- Sponsorship — supporting ongoing open-source contribution work
- Contract/freelance work — Python, graph databases, ML infrastructure, dev tooling
- Interesting problems — always happy to talk shop
📍 California · 💬 Reach out via GitHub
If my contributions have helped your project, consider sponsoring my work.