Research, Safety

Thinking Machines Lab Inc.; San Francisco (CA); On-site; USD 350,000 - 475,000; Full time

Benefits offered by this job

Health, dental, and vision benefits

Unlimited PTO

Parental leave

Relocation support

Job summary

Thinking Machines Lab Inc. in San Francisco is seeking a safety researcher to bridge research and hands-on engineering, focusing on making models safe and trustworthy.

You will explore how training shapes refusals and how to evaluate boundaries, designing experiments to inform model training and evaluation strategies.

Join a team that values rigorous analysis, data-driven evaluation, and responsible AI practices, with opportunities to influence real-world deployments.

Qualifications

• Bachelor’s degree or equivalent in Computer Science, Machine Learning, Physics, Mathematics, or related field with strong theoretical and empirical grounding.
• Background in AI safety research with hands-on experience in at least one safety area (RLHF/RLAIF, alignment, red-teaming, etc.).
• Proficiency in Python and familiarity with DL frameworks; comfortable debugging distributed training.

Responsibilities

• Work across data curation, safety-focused fine-tuning, evaluations, and red-teaming across the development stack.
• Design experiments to understand how training shapes model refusals and safe behavior.
• Develop safety evaluations and synthetic data to test boundaries and risk surfaces.
• Collaborate with teams to implement mitigations for safety failures and jailbreaks.

Skills

Python programming

Deep learning frameworks

Communication clarity

AI safety research background

Education

Bachelor’s degree or equivalent in CS/ML/Physics/Math

PhD in CS/ML/Physics/Math optional

Tools

PyTorch

TensorFlow

JAX

Job description

As a safety researcher, you'll work toward ensuring our models are safe and trustworthy. The role sits at the intersection of research and hands-on technical work. A central question is how models come to handle harmful or dual-use requests: what they learn from data, how training shapes where they refuse and where they engage, and what makes those boundaries reliable. You'll explore the science behind these behaviors and design experiments that inform how our models are trained and evaluated.

What You’ll Do

We are hiring across the entire development stack — from pre-training data curation to safety-focused fine-tuning, evaluations, and red-teaming. During project selection we’ll take into account your interests and experience alongside organizational needs. This flexible approach allows us to match talented safety researchers with the teams where they'll have the greatest impact and growth potential.

Here are example areas you may contribute to depending on your area of expertise and interest:

• Build data filtering pipelines and quality classifiers to shape what models learn from pre-training corpora, and study how those early interventions affect downstream safety behavior.
• Apply post-training techniques, including RL from human and AI feedback and policy-based reasoning approaches, to shape how models handle harmful, sensitive, and dual-use requests.
• Design, build, and maintain safety evaluations, with particular focus on measuring model behavior on long-horizon and agentic tasks.
• Generate and curate synthetic data to train and evaluate models on refusal boundaries and safety-relevant behaviors.
• Red-team our models and products to surface failure modes, jailbreaks, and emergent risks before deployment, and design mitigations for what you find.

Skills and Qualifications

Required qualifications:

• Bachelor’s degree or equivalent experience in Computer Science, Machine Learning, Physics, Mathematics, or a related discipline with strong theoretical and empirical grounding.
• Background in AI safety research, with hands-on experience in at least one area of safety, such as: RLHF/RLAIF, alignment and preference modeling, deliberative alignment, safety evaluations, or red-teaming.
• Proficiency in Python and familiarity with deep learning frameworks (e.g., PyTorch, TensorFlow, or JAX). Comfort debugging distributed training and writing code that scales.
• Clarity in communication, an ability to explain complex technical concepts in writing.

Preferred qualifications — we encourage you to apply if you meet some but not all of these:

• Experience building evaluations for long-horizon, multi-step, or agentic tasks.
• Experience generating synthetic data at scale for training or evaluation.
• Experience with modern red-teaming/jailbreaking techniques.
• Research contributions in AI safety — publications, open-source evaluations, or public red-teaming work.
• Familiarity with the AI safety literature and current open problems (e.g., scalable oversight, reward hacking, jailbreak robustness).
• PhD in Computer Science, Machine Learning, Physics, Mathematics, or a related discipline with strong theoretical and empirical grounding; or, equivalent industry research experience.

Logistics

• Location: This role is based in San Francisco, California.
• Compensation: Depending on background, skills and experience, the expected annual salary range for this position is $350,000 - $475,000 USD.
• Visa sponsorship: We sponsor visas. While we can't guarantee success for every candidate or role, if you're the right fit, we're committed to working through the visa process together.
• Benefits: Thinking Machines offers generous health, dental, and vision benefits, unlimited PTO, paid parental leave, and relocation support as needed.

As set forth in Thinking Machines' Equal Employment Opportunity policy, we do not discriminate on the basis of any protected group status under any applicable law.