ByteDance
Research Scientist Graduate (Seed Model-Alignment&Preparedness) - 2027 Start (PhD)
San Jose, California · Full-time · Posted Oct 28, 2026
- Experience
- No years specified; completing or recently completed a PhD. Relevant research or engineering experience is required.
- Education
- PhD in Computer Science, Electrical Engineering, Electrical and Computer Engineering, Physics, Mathematics, or a related discipline.
About the job
Research and develop methods to make frontier AI models and agents safer, focusing on agent reinforcement learning, risk assessment, automated red teaming, oversight, and deployment.
What you'll do
- Train safer agents using RL, SFT, and preference optimization; design environments, rewards, and data to reduce unsafe behavior.
- Assess frontier risks and build automated red teaming, monitoring, oversight, and permission-control methods.
- Analyze failures, run controlled experiments, and collaborate across teams to deploy effective safeguards at scale.
What they ask for
- Completing or recently completed a PhD in a listed or related discipline.
- Strong foundations in ML, deep learning, and statistics, plus research or engineering experience in LLMs, RL, or related fields.
- Proficiency in Python and PyTorch (or similar), including implementing algorithms and building and debugging experiment pipelines.
- Hands-on experience in at least one of agent RL, model training, dataset development, agent systems, evaluation, or ML infrastructure.
Nice to have
- Experience with agent RL, SFT, RLHF/RLAIF/RLVR, distributed training, or automated red teaming.
- Research or experience in AI control, scalable oversight, CoT monitoring, mechanistic interpretability, prompt injection defense, or high-stakes risk mitigation.
Tools
Python, PyTorch, RL, SFT, RLHF/RLAIF/RLVR
AI focus
AI safety research and development, especially agent RL, alignment, safety evaluation, preparedness, red teaming, monitoring, and oversight.
Model-written summary · the original posting is authoritative