About the Role:

The bottleneck in building reliable and useful AI systems isn’t writing code anymore. It’s knowing what good looks like, applying that knowledge and experience from production usage into building rigorous evals, and using those evals to guide continuous product self-improvement loops.

As the Tech Lead Manager of AI Evaluations, you will lead a team of applied AI scientists and engineers and build the evaluation strategy and platform that helps Glean ship delightful and reliable AI experiences. You will make evaluation a continuous part of how we develop and operate models, harnesses, assistants, agents, and other AI products at Glean.

You’ll guide the team in translating emerging AI/ML research into reliable evaluation systems, build scalable, efficient and reusable eval infrastructure and trusted quality signals, and work across R&D, product, and marketing teams to turn evaluation results into better products and customer outcomes.

You will:

• Own the AI evaluation eng roadmap and technical direction end to end: from defining what success means and building the evaluation platform to integrating results into release and product decisions.
• Build scalable, reusable evaluation infrastructure for datasets, experiments, graders, trace capture and replay, result analysis, and dashboards. Make evaluations reproducible, versioned, cost-aware, and easy for product and engineering teams to use.
• Create and maintain representative evaluation sets from expert-curated examples, appropriately governed production signals, synthetic cases, and historical failures. Cover common workflows, edge cases, long-tail requests, adversarial inputs, and important user and domain slices; protect held-out acceptance sets from tuning leakage.
• Define a balanced quality measurement system for AI products, including correctness, groundedness, retrieval and context relevance, completeness, instruction following, citation quality, user preference, and task success. For agents, evaluate tool selection and arguments, multi-step trajectories, policy compliance, recovery, and final task state, not just the final response.
• Develop a fit-for-purpose mix of deterministic checks, model-based graders (including LLM-based autograders), and human evaluation. Establish clear rubrics; validate and calibrate automated judges against human judgments; and measure evaluator reliability and limitations so teams can trust and interpret scores.
• Integrate evaluation into development and operations: fast CI regression checks, broader pre-release suites, model and prompt comparisons, and sampled production evaluation. Build feedback loops from user feedback and production traces into new test cases, and use monitoring, canaries, and experiments to detect regressions and drift.
• Make quality tradeoffs measurable by connecting evaluation outcomes to latency, reliability, and cost. Help teams diagnose failures by slice and trace, prioritize improvements, and make evidence-based decisions about models, routing, prompts, and system changes.
• Build and lead a high-performing team of applied AI scientists and engineers; coach and develop team members. Partner closely with product, infrastructure, security, and application teams to align evaluation standards and deliver the highest-impact work.

About you:

• 10+ years of experience in software engineering, ML infrastructure, data systems, or a related technical field. Experience must include building and operating production systems.
• 5+ years of engineering or applied AI team management experience, with a track record of technical leadership, developing engineers and scientists, and delivering cross-functional projects.
• Hands-on experience designing, building, or operating evaluation systems for LLM/GenAI products, including model-based evaluators or autograders; practical understanding of how to assess model, retrieval/RAG, assistant, or agent behavior.
• Familiarity with evaluation methods and tradeoffs: curated and production-derived datasets, benchmark creation, deterministic metrics, rubric-based and pairwise judgments, human review, judge validation and calibration, human-model alignment, and statistical analysis of results.
• Understanding of agent evaluation, including tool calling, multi-turn behavior, trajectory quality, task completion, and recovery from errors; experience with safety testing or red teaming is a plus.
• Strong coding ability in Python or another production language, with sound engineering fundamentals in testing, APIs, data modeling, and reliable distributed systems.
• Experience building scalable data and developer platforms, experiment pipelines, or CI/CD integrations; familiarity with cloud infrastructure, Kubernetes, data stores, and observability is beneficial.
• Strong judgment about measurement quality, including dataset representativeness, judge calibration, reproducibility, uncertainty, and the risk of optimizing to a narrow metric.
• Knowledge of public benchmarks across domains and capabilities is a plus.
• Product-minded and collaborative: able to translate ambiguous quality goals into clear measures, communicate findings to technical and non-technical audiences, and influence roadmaps across teams.
• MS/PhD in Computer Science, AI or a related field.