TLDR
Worked across field engineering and developer experience for an enterprise AI observability platform.
I shipped production integrations for agent frameworks, worked with Microsoft’s ASSERT team on behavioral-eval observability, overhauled documentation, and built internal Cursor Automations for documentation and GTM work.
The Problem
Agent frameworks were changing quickly, but enterprise teams still needed observability integrations they could use in production.
The documentation had also become a support pain point. Customer-facing teams often had to explain workflows live because the written guidance was not enough.
Internally, documentation upkeep and lead research involved repetitive work that could be prepared by an agent and reviewed by a person.
What I Shipped
Production observability integrations
Built integrations across Google ADK, AWS Strands, and other agent frameworks. I also worked with Microsoft’s ASSERT team as an early partner on behavioral-eval observability.
Cursor SDK cookbook
Published an observability cookbook within hours of the Cursor SDK public beta. It gave developers a working path for tracing SDK-built agents.
Documentation overhaul
Reworked the documentation around the questions customers actually asked. It went from a major support pain point to guidance customer-facing teams could confidently share during calls.
Cursor Automations now monitor code changes and prepare documentation updates for review.
GTM Automation
Built a Cursor Automation that collects lead signals, drafts outreach, and sends it to Slack for the GTM team to review.
What I Learned
An API reference alone is rarely enough when you’re connecting two fast-moving systems. People need a working example they can understand and adapt.
Better docs cut down on repeated explanations on calls. That helped customers and the team supporting them.
Related
- about: More about my work
- AI-Ready Docs for Ragas: Eval cookbooks and docs automation for an open-source framework