AnthropicSan Francisco, CAConduct empirical research on AI's economic impacts, labor markets, and societal effects using external data and Python. Produce public outputs under mentorship, focusing on AI safety and policy recommendations; 4-month full-time program open to varying experience levels.
Conducts research in mechanistic interpretability to understand deep network representations and enhance AI safety. Requires PhD or equivalent in ML/CS, 2+ years research engineering with Python, and passion for safe AGI.
Advances AI products through post-training SOTA LLMs using supervised and reinforcement learning techniques on rich query datasets. Owns data pipelines, training frameworks, and model integration while collaborating across teams. Requires 2-6+ years in large-scale LLMs and Python/PyTorch expertise; PhD preferred.
Develops and deploys state-of-the-art ML models and agents for enterprise GenAI using RL training and post-training algorithms. Requires 1-3 years LLM production experience, RLHF expertise, recent top publications, and advanced CS degree.
Develops and optimizes post-training algorithms for agent RL platforms, focusing on LLM training, inference frameworks, and multi-agent systems. Requires 1-3 years production LLM experience, expertise in PyTorch/CUDA, RLHF/PPO, and advanced degree.
AI Research Resident collaborates on research projects developing benchmarks and environments for long-horizon AI agents, identifying model failure modes, and training autonomous agents. Requires current MS/PhD enrollment, RL experience, systems engineering, and strong publications.