I work on the parts of a system that are easy to claim and hard to prove — latency you have to measure at the tail, permissions you have to enforce rather than document, a wrong number that still renders as a working page. Lately that has mostly meant fixing other people's production code: 79 pull requests across 34 organisations — 29 merged upstream, 32 in review.
Where a number appears below, it was measured, and the method is written down in that project's README.
|
Merged upstream
|
In review now
|
Currently — in onecli/onecli: #626 fix(gitlab): cover GET /api/v4/user with a read tool so whoami isn't blocked, opened 2026-10-26.
Also around:
- rl-puzzle-agents — PPO, GRPO and DQN benchmarked head to head in a custom Gymnasium environment
- research-scout — arXiv retrieval agent with local embedding rerank and source-linked digests
- llm-peer-review-agent — three-round self-critiquing reviewer scored against a single-pass baseline
2025 |
Barclays — CISO Office | data-quality and deployment-automation systems |
2025 |
$10,000 · Stellar Development Foundation | Kryo |
2025 |
Top 14 of 143 · Tree Canopy Segmentation Challenge | weighted mAP 0.33 → 0.42 |
2024 |
$10,000 · Emergent Ventures | algae-based air purifier |
2024 |
Top 3 · BCG Ideathon | quantitative microfinance credit model |
2024 |
2nd place · Plaksha Hackathon | |
2024 |
ISC2 Certified in Cybersecurity |
Contribution counts, panels and the lists above are rebuilt from the GitHub API every six hours. Output last changed 2026-10-27.

