Teacher

• Guide 25–30 participants in a 14-week AI safety program by leading Saturday sessions and providing remote support.
• Lead weekly 1-hour lectures on advanced ML topics including RL, transformers, mechanistic interpretability, and model evaluation.
• Deliver 4pm Saturday sessions on AI safety ecosystems, risks, and current research.
• Provide pair programming support and asynchronous Slack help for technical problems and environment setup.
• Review project proposals and provide technical feedback during the final curriculum phase.

This role directly contributes to building a skilled workforce in AI safety, helping to mitigate risks from advanced AI systems and ensuring responsible AI development.

• Expertise in advanced machine learning topics such as reinforcement learning, transformers, mechanistic interpretability, and model evaluation.
• Experience teaching or mentoring in AI safety or related technical fields.
• Strong communication and remote collaboration skills.
• Ability to provide technical feedback on project proposals.

Compensation and benefits not specified in the posting.

The Technical Alignment Research Accelerator is an organization focused on advancing AI safety research and training the next generation of alignment researchers.