I make hard AI infrastructure feel obvious, for agents and for the developers who build them.

  Website         LinkedIn         X         YouTube         Email

I lead Agent Experience and Developer Experience at Blaxel, now part of Baseten, and Developer Relations at Baseten.

Blaxel gives AI agents fast, isolated sandboxes to work in. Baseten runs inference for production AI. My job is to make that whole surface feel obvious: the CLI, SDKs, docs, integrations and workflows that coding agents can operate directly, and that developers can ship with on day one.

• Agent experience. Designing the CLI, SDK, skill and MCP surfaces so agents like Claude Code, Codex and Cursor can drive Blaxel end to end without a human translating in between.
• Developer experience. Owning the path from the first bl command to production: onboarding, docs, examples, error messages and integrations with the frameworks developers already use.
• Developer relations. Tutorials, cookbooks, launches, talks and the feedback loop from the community back into the product, across Baseten and Blaxel.

I am most useful when the problem is ambiguous, the surface area is large, and someone needs to own the whole path from product shape to implementation to adoption.

Give your coding agent the Blaxel skills, then ask it to spin up a sandbox:

npx skills add blaxel-ai/agent-skills

Works with Claude Code, Codex, Cursor and most agents that read skills. Prefer to drive it yourself? Install the CLI with brew install blaxel-ai/blaxel/blaxel and follow the quickstart.

SystemSculpt AI is my hobby project: an agent workspace for Obsidian, with 37k+ downloads and 190+ GitHub stars.

Building agent infrastructure, developer tools, or something slightly impossible? Let's talk.