Mercor
Research Engineer - Environments, Data and Post-Training
$180K–$500K
San Francisco · On-site · Full-time · Posted Mar 25, 2026
About the job
Build and operate post-training, synthetic-data, and evaluation systems that improve frontier language models’ tool use, agentic behavior, and real-world reasoning.
What you'll do
- Run post-training and RLVR experiments, including reward-shaping and algorithmic improvements.
- Build scalable data generation and augmentation pipelines, and assess data quality and benchmark performance.
- Create and operate rubrics, evaluators, benchmarks, and scoring systems for LLM training and evaluation.
What they ask for
- Strong applied research background focused on post-training and/or model evaluation.
- Strong coding proficiency and hands-on experience with machine learning models.
- Strong understanding of data structures, algorithms, backend systems, and core engineering fundamentals.
- Familiarity with APIs, SQL/NoSQL databases, and cloud platforms.
Nice to have
- Real-world post-training team experience in industry.
- Publications at top-tier conferences such as NeurIPS, ICML, or ACL.
Pay
$180K – $500K USD base salary
Tools
GRPO, DAPO, APIs, SQL/NoSQL databases
AI focus
Post-training and RLVR, reward shaping, synthetic data generation, and large-scale LLM evaluation to improve tool use, agentic behavior, and real-world reasoning.
Model-written summary · the original posting is authoritative