TrainLoop

Training infrastructure for long horizon agents

Research Scientist

$150K - $250K•0.50% - 1.50%•San Francisco, CA, US
Job type
Full-time
Role
Science, Research
Experience
3+ years
Visa
US citizen/visa only
Connect directly with founders of the best YC-funded startups.
Apply to role ›

About the role

Develop new methods for building specialized models that learn reliably and solve difficult tasks. We’re hiring research scientists across post-training, reinforcement learning, continual learning, multimodal models, evaluation, and interpretability. Your research focus will depend on your strengths and the questions you pursue with the team.

You might develop new training methods, investigate how models acquire and retain skills, design evaluations for long-horizon reasoning, study model internals, or explore architectures and learning strategies for specialized domains. You’ll formulate hypotheses, implement and run experiments, compare against meaningful baselines, and work with research engineers to turn useful findings into practical capabilities. Communicate results clearly, including limitations and competing explanations.

Requirements

• Typically 3+ years of relevant ML research experience, or equivalent demonstrated research contributions. Relevant doctoral research counts toward experience.

• Independent drive for research. Top candidates can discuss at length contemporary research in language model training.

• At least one substantial original research contribution demonstrated through a paper or open-source implementation.

• Depth in at least one of post-training, reinforcement learning, continual learning, multimodal learning, evaluation, interpretability, systems, or inference.

• For work focused on model training: typically 1+ year of hands-on language or multimodal model training, including doctoral research, or equivalent completed training projects.

About TrainLoop

TrainLoop
Founded:2025
Batch:W25
Team Size:12
Status:
Active
Location:San Francisco
Founders