Co-op Researcher – LLM Post-Training & Agentic Coding
Core
Conduct cutting-edge research on LLM post-training, reinforcement learning, reasoning, agentic coding, and self-improving agents, taking technically challenging ideas from initial hypothesis to rigorous experimental validation.
Role type
Researcher (LLM Post-Training & Agentic Coding)
Builds
Intelligent agents capable of reasoning, tool use, code generation, and solving complex real-world tasks; scalable infrastructure for LLM training, inference, and synthetic data generation.
Domain
Artificial Intelligence / Machine Learning / Large Language Models
Deliverable
production ML models
Required skills
LLM post-training techniques, reinforcement learning, reasoning, agentic coding, Python, PyTorch, distributed systems, ML infrastructure, experimental design, large-scale evaluation
Preferred skills
Transformer-based models, supervised fine-tuning, preference optimization, distillation, synthetic-data-driven training, vLLM, SGLang, FSDP, DeepSpeed, Hugging Face, Docker, Linux, Git, multi-GPU compute environments
Responsibilities
Develop new approaches for agent training, task and data synthesis, environment generation, and agent evaluation; Build and evaluate intelligent agents; Build reliable and scalable infrastructure for LLM training and inference; Improve efficiency and reliability of GPU-based post-training and inference workflows; Own research problems end-to-end from literature review to communicating findings.
Seniority
Student Co-op (Senior Bachelor's, Master's, or PhD)