Partner directly with the Applied Research Team Lead to define and evolve team process, technical standards, and the quality bar for agent behavior.
Take real user experience and data through the full arc, from concept to a reliable agent design to a shipped, production-grade product.
Own or share ownership of the standing function of assessing AI tools, models, and providers for reliability and fit.
Raise the technical bar across a team spanning research-leaning grad students to engineers newer to AI, and turn personal expertise into shared tooling and standards the whole team can use.
Operate pragmatically inside current infrastructure constraints while flagging where structural fixes are needed.

Strong applied research background (NLP/ML) with real production experience, shipped systems, not only research output.
Fluency in the modern stack: context engineering, agent design and orchestration, and evaluation of LLM-based systems.
An execution and operations mindset, biased to ship, comfortable defining good enough versus chasing optimal.
A track record of building standards or tooling that lifts a whole team, or clear readiness to grow into that.
Comfortable owning ambiguous problems end to end with minimal oversight.