The opportunity
Code Intelligence sits at the intersection of agentic engineering and product, running agents against real customer source code in production. This is a chance to own an AI-native team at the center of a business-critical growth area.
What you'll do
Lead and develop an existing engineering team, building trust, setting: technical direction, and establishing a high bar for ownership and execution.
Own the roadmap and execution across four areas: source code indexing, PR Agentic Reviewers (agents that generate artifacts like metric descriptions on customer pull requests), shift-left work (extending code-writing agents to make direct changes to customer repos), and closed-loop feedback and evaluation systems including LLM-as-judge.
Set the team's product strategy independently while partnering day-to-day: with product teams company-wide — this team functions as its own product org.
Build and operate scaled LLM and agent software in production: LLM APIs, agent frameworks, harness and tool-use layers, prompt and eval tooling, and closed-loop evaluation systems that measure and improve agent output quality.
Partner closely with customers and internal stakeholders to deeply understand: needs and translate them into technical direction.
Drive the team's AI-native development practices, setting high standards for: safety, validation, and increasing agent autonomy over time.
What they're looking for
- Experienced engineering manager with a track record of shipping products with: direct customer interaction, not just internal stakeholders.
- Hands-on experience building scaled LLM software: LLM APIs, agent frameworks, prompt and eval tooling, harness and tool-use development, and closed-loop evaluation systems including LLM-as-judge.
- Strong product and customer mindset: you put solving customer problems first and know how to ship and iterate quickly.
- Comfortable setting your own product strategy in an ambiguous environment: rather than executing a pre-defined roadmap.
