Research Scientist, AI

What you'll do

• Develop model training and post-training approaches, including research in reinforcement learning, synthetic data, reasoning, and agentic AI.
• Lead end-to-end research initiatives and develop benchmarks and evaluations for reasoning, robustness, and generalization.
• Partner with engineering to bring research into production and improve system performance and reliability.

What they ask for

• Track record of high-impact research, including projects successfully deployed to advance a product, technology, or organization.
• Hands-on experience in foundation model training or post-training, including at least one of large-scale pretraining, reinforcement learning, reasoning-focused training, or synthetic data generation.
• Proficiency in Python and modern ML frameworks such as PyTorch or JAX, with experience implementing and scaling training experiments.
• Research judgment, experimental rigor, and technical leadership, including driving initiatives and mentoring scientists.

Nice to have

• Experience with frontier-scale models, distributed training, large-scale reinforcement learning, or inference-time scaling.
• Research in neuro-symbolic AI, autoformalization, automated theorem proving, program synthesis, or formal verification; or a strong record of influential publications or deployed breakthroughs.

Tools

Python, PyTorch, JAX, Lean 4

AI focus

Research on training and post-training models to formalize complex domain knowledge and reason over symbolic world models, including reinforcement learning, synthetic data generation, agentic AI, and evaluation.