Lead a team of engineers, researchers, and data scientists, performing QA on AI agent outputs
Evaluate RAG pipeline architectures, challenge prompt engineering decisions, and review system designs
Build relationships with internal business teams, understanding workflows and translating pain points into actionable technical work
Set priorities and decide what the team works on across both AI Enablement and AI Experiments tracks
Serve as the primary liaison with internal business teams, gather requirements, and validate priorities
Communicate progress clearly to leadership and identify new internal processes where AI could reduce effort
Guide the team’s technology roadmap, evaluate architecture decisions and optimize AI infrastructure spend

8+ years of experience in technical product management, engineering management, or a similar role where you’ve led technical teams building AI/ML or data-driven products
Strong understanding of LLM-based systems you know how RAG pipelines work (retrieval, embedding, re-ranking, context window management), how prompt engineering affects output quality, and how agent orchestration patterns handle multi-step workflows.
Comfortable with Python you can read Python code, understand what it does, review pull requests for logic and architecture (not just style), and write quick scripts when needed to test an idea or validate data.
Familiar with Azure cloud infrastructure you understand how AI workloads are deployed, monitored, and scaled in Azure (or equivalent cloud platforms). You can have an informed conversation about compute costs, model serving, and infrastructure architecture.
You understand the cost structure of LLM-based systems token economics, model pricing tiers, the cost implications of different prompt lengths, caching strategies, and model routing approaches. You can optimize AI spend at a technical level, not just a budget level.
You can evaluate AI outputs critically you know what a hallucination looks like, you understand why an agent might produce inconsistent results, and you can diagnose whether the issue is in the prompt, the retrieval layer, the model choice, or the data.

Senior AI Technical Product Manager

Key skills