Building AI agent tooling, browser automation workflows, and reliable systems.
- Coding-agent orchestration and developer workflows
- Browser automation for real maintainer and agent tasks
- OSS fixes, regression prevention, and cross-platform reliability
Coding-agent orchestration service with a Django backend, an asyncio orchestrator, and an Angular dashboard.
Turn screenshots into actionable information with a workflow designed for fast analysis.
Image-generation service scaffold for Colab GPU runtimes with API exposure and runtime operations scripts.
Public Apache-2.0 browser automation CLI for AI agents.
Recent merged work:
- PR #734: aligned the Windows native daemon port hash with the client
- PR #731: fixed
newTab()behavior in persistent-context mode
I care about the kind of OSS work that keeps projects usable: bug reproduction, targeted fixes, regression coverage, and maintainable workflows.
I prefer structure over hype:
- clear interfaces
- testable behavior
- reproducible workflows
- systems that stay useful under uncertainty
If you are working on AI agents, browser automation, or maintainer tooling, feel free to reach out through GitHub issues, discussions, or pull requests.



