Mercor is at the intersection of labor markets and AI research, partnering with leading AI labs and enterprises to provide the human intelligence essential to AI development. They are looking to train Large Language Models to master tool-use, agentic behavior, and real-world reasoning.
Requirements
- Strong programming skills in Python, Go, or JavaScript, with an ability to write clean, reliable, production-grade code.
- Understanding of data structures, algorithms, backend systems, and core engineering fundamentals.
- Familiarity with APIs, SQL/NoSQL databases, and cloud platforms.
- Experience training models or evaluating model performance
Responsibilities
- Work on post-training and RLVR pipelines to help Mercor understand how datasets impact model performance.
- Design and run reward-shaping experiments and algorithmic improvements (e.g., GRPO, DAPO) to improve LLM tool-use, agentic behavior, and real-world reasoning.
- Quantify data usability, quality, and uplift on key benchmarks.
- Build data generation and augmentation pipelines that scale with training needs.
- Create and refine rubrics, evaluators, and scoring frameworks that push the boundaries of what LLMs can learn.
- Collaborate closely with research engineers, applied AI teams, and experts producing data.
- Operate in a fast-paced, experimental research environment with rapid iteration cycles.
Other
- Pursuing a degree in Computer Science or a related field (graduating 2025–2027); ability to start in early 2026 is strongly preferred.
- Curiosity and passion for AI research, reinforcement learning, and fast-moving startups.
- Excitement to work in person and thrive in a high-intensity, high-ownership engineering environment.
- Work samples, artifacts, or code repositories demonstrating relevant skills
- Publications at ACL, NeurIPS, or ICML conferences
