About the role#
As a Research Scientist Intern at Mercor, you will contribute to research on post-training, reinforcement learning with verifiable rewards (RLVR), data generation, and model evaluation. You will investigate how datasets, rewards, and training methods influence the capabilities and behavior of large language models. You will work with research scientists, engineers, and domain experts to turn open-ended questions into rigorous experiments that support our research agenda, benchmark releases, and frontier AI systems.
What you'll need#
- Currently pursuing a master’s or PhD in computer science, machine learning, statistics, mathematics, or a relevant field.
- Demonstrated ability to formulate research questions, design experiments, and draw sound conclusions from empirical results.
- Experience in training, fine-tuning, or evaluating language models, agentic AI systems, or RL environments.
- Experience developing benchmarks, evaluation methodologies, or data-quality measures.
- At least one publication or open-source project.
- Strong programming skills in Python and the ability to write reliable research code.
- Familiarity with machine learning fundamentals, experimental design, and statistical analysis.
- Intellectual curiosity and comfort operating in an environment with rapid iteration and high ownership.
What you'll do#
- Develop and investigate research questions regarding post-training, RLVR, data quality, and model evaluation.
- Design and run controlled experiments to understand how datasets, rewards, and training strategies impact model performance.
- Study reward-shaping and post-training methods, including GRPO and DAPO.
- Build research tooling and data pipelines to conduct experiments at scale.
- Develop methods to measure data quality, usability, and performance uplift on benchmarks.
- Design and evaluate datasets, rubrics, and scoring frameworks for complex model capabilities.
- Conduct systematic error analysis to identify failure modes and improvement opportunities.
- Analyze results and communicate findings through reports, research artifacts, and presentations.
- Collaborate with research scientists, engineers, and applied AI teams.
- Contribute to research publications, benchmark releases, and other public outputs.
Location & details#
- Location: San Francisco, California, United States.
- Modality: On-site, five days a week.
- Employment: Full-time, paid internship at $80 per hour.
- Term: Rolling.
- Benefits: Mentorship from experienced researchers, $1,500 monthly meal stipend, $200 monthly laundry reimbursement, $200 monthly wellness reimbursement, and a free Equinox membership.


