Research Engineer, QC Automation

About the Role

An early-stage AI infrastructure company is hiring a Research Engineer, QC Automation - the #1 priority hire on the engineering team right now. You'll own end-to-end automation of quality control for AI training data generated by companies using the platform's infrastructure. This is a high-impact, high-autonomy role sitting at the intersection of data engineering, research, and systems design.

You'll be joining a ~15-person engineering group composed of Olympiad medalists, AI startup founders, and published researchers, working on one of the most critical challenges in post-training data quality for reinforcement learning.

What You'll Do

Automate quality control for training data produced by companies using the platform's infrastructure.

Build QC systems grounded in true understanding and human judgment - not heavy reliance on LLMs.

Define and enforce quality standards for post-training datasets.

Design experiments and metrics to grade agent outputs.

Partner with data vendors to debug quality issues, diagnose agent failure modes, and improve data generation processes.

Translate QC learnings into auditing systems, including sampling strategies and rule-based or model-assisted validation pipelines.

Continuously integrate QC learnings into infrastructure tooling and the data vendor portal to reduce anomalies, inconsistencies, and edge cases.

What We're Looking For

Required:

2–4 years of experience in engineering or research roles.

Proficiency in Python, Docker, and Linux environments.

Strong understanding of what "good data" means and how to measure it.

Proven experience building scalable data validation pipelines and automated QA/QC systems end-to-end.

Experience working on benchmarks and evals - including reasoning about realistic tasks, reliable rubrics, and useful trajectories for RL training.

Knowledge of statistics and comfort designing metrics, experiments, and QA/QC processes.

Strong written and verbal communication skills for collaborating across time zones.

Genuine curiosity across domains and an ability to ask questions that drive understanding.

Ability to thrive in unstructured problem spaces and work independently in a fast-paced, early-stage startup environment.

Nice to have:

Background in AI evaluation, reinforcement learning environments, or post-training data pipelines.

Experience with reward signal analysis or reward hacking detection.

Prior startup experience or demonstrated comfort with ambiguity and self-direction.

Compensation & Benefits

Salary: $150,000 – $250,000 USD annually

Visa sponsorship available for eligible candidates

Location

San Francisco, CA (on-site) for U.S.-based candidates

Singapore (on-site) for Southeast Asia–based candidates

Fully remote as an independent contractor for candidates based elsewhere, particularly in Europe