Senior ML Engineer (AI Research)
Core
Conducting experiments to figure out efficient ways to train large language models on traces of interactions, exploring guided generation and search in trajectory space, and mining web-scale data for model post-training.
Role type
Senior IC machine-learning engineer (AI research)
Builds
Research findings applied in practice by adjacent teams for agentic systems and reasoning models
Domain
AI research, reinforcement learning, large language models
Deliverable
research
Required skills
theoretical foundations of machine learning and reinforcement learning, modern deep learning for language processing and generation, training large models on multiple computational nodes, software engineering in Python, modern deep learning frameworks (JAX), designing and analyzing machine learning experiments with statistical rigor, formulating research questions and designing experiments
Preferred skills
deep reinforcement learning for LLMs (reward modeling, DPO, PPO), familiarity with LLM ideas (RoPE, ZeRO/FSDP, Flash Attention, quantization), building and delivering products in a startup-like environment, engineering complex distributed data processing systems or high-load web services, open-source projects
Technologies
Python, JAX
Responsibilities
Conduct experiments to train large language models on interaction traces, explore guided generation and search methods in trajectory space, mine relevant data at web scale for model post-training, conduct experiments with reinforcement learning configurations in verifiable domains, explore methods to train AI agents on tasks with non-verifiable reward signals
Seniority
Senior, hands-on IC
