Research, Safety

Sourced from zerogtalent.com

Thinking Machines Lab is hiring an AI Safety Researcher to ensure their frontier models are safe, trustworthy, and aligned with human values. This role combines hands-on technical work with research across the entire model development stack, from pre-training data curation to safety-focused fine-tuning, evaluations, and red-teaming. You will design experiments, build safety pipelines, and develop mitigations for emergent risks before deployment.

• Build data filtering pipelines and quality classifiers for pre-training corpora
• Apply post-training techniques like RLHF/RLAIF to shape model behavior on harmful requests
• Design and maintain safety evaluations for long-horizon and agentic tasks
• Generate synthetic data to train models on refusal boundaries
• Red-team models to surface failure modes, jailbreaks, and emergent risks

• Bachelor’s degree or equivalent in CS, ML, Physics, Math, or related field
• Hands-on experience in AI safety research (e.g., RLHF/RLAIF, alignment, red-teaming)
• Proficiency in Python and familiarity with PyTorch, TensorFlow, or JAX
• Ability to debug distributed training and write scalable code
• Strong written communication skills