CareerPlanSign in

Engineering Manager, Evals

San Francisco💼 Full-time🗓 2026-06-03 → 2026-09-26

Core

Lead the team creating high-signal evaluation datasets for coding agents and building tools for engineers to write and run evaluations.

Role type

Engineering Manager, Evals

Builds

High-signal evaluation datasets (e.g., CursorBench), developer-friendly tools for running evals, and online evaluation systems.

Domain

AI/ML, Coding Agents, Software Engineering

Deliverable

production ML models

Required skills

People leadership and coaching, aligning research/product/data/infrastructure, defining quality signals and metrics, operating evaluation or measurement systems, data acumen, understanding model and agent behaviors.

Preferred skills

Experience shipping production systems, building evaluation systems (AI evals, experimentation platforms, ranking/relevance, search quality, reliability instrumentation).

Technologies

CursorBench, online evaluation systems

Responsibilities

Set the eval roadmap end-to-end, lead and grow a team of engineers and researchers, guide the evolution of evaluation benchmarks, define online quality signals and guardrails, integrate evals into launch and model training decisions.

Seniority

Manager, hands-on leadership

Sourced via ashby · Listed on CareerPlan, which tracks 70,000+ jobs from 20+ sources.