- Own AI tooling adoption across π: Identify where AI tools can improve velocity, build or integrate the right solutions, teach teams how to use them, and drive adoption.
- Build internal AI tooling and integrations: Build backend services, scripts, workflows, user interfaces, LLM integrations, and agent infrastructure.
- Make AI agents ergonomic: Own workflows for cloud agents, agent management, and internal automation that are easy to use, easy to monitor, and easy to trust.
- Build tools for engineering, research, and operational velocity: Help engineers use AI to write, test, debug, review, and validate code faster. Empower researchers to extract signals and iterate quickly and confidently. Work with operations and recruiting to understand their workflows and build tools that give them leverage.
- Own best practices and enablement: Create playbooks, examples, onboarding, office hours, demos, and shared workflows that help people learn from the best AI users at π.
- Partner on security and data access: Ensure AI tools have the right access to be useful while respecting data boundaries, permissions, and company policies.
- Evaluate build vs. buy: Maintain a strong perspective on the AI tooling ecosystem, evaluate commercial tools, and recommend what π should adopt.
- Measure impact: Define success metrics for adoption, productivity, and satisfaction. Use feedback and data to understand what is working, what is not, and where to invest.