AI Researcher – AI Chemistry

You've trained and post-trained large models yourself and you want them to change the physical world. Build the models behind AI-driven chemistry.

A Series B company in San Francisco combines AI, its own labs and a manufacturing network to make chemicals for more than 100 enterprise customers. It is now building its founding AI team. You'll create the loop that turns chemistry data and messy lab results into dependable models for planning how molecules are made: continued pretraining, fine-tuning, synthetic data, distillation, benchmarks and distributed training, all the way to models chemists use every day. You don't need a chemistry degree; what you've achieved counts more than your path.

What you'll own

• Training and post-training runs, including reinforcement learning from experimental results
• Synthetic data and distillation pipelines with clear quality controls
• Evaluations, ablations and baselines you can trust
• Taking open research questions all the way to deployed models, together with chemists and product engineers

Must-haves

• You've run large-model training or post-training yourself and were responsible for the experiments, not just helping on someone else's
• Practical experience training across many GPUs with FSDP, DeepSpeed or Megatron, using PyTorch or JAX
• Specialist depth in one or more of: reinforcement learning and reasoning, training LLMs at scale, synthetic data and distillation, world models and planning
• 5+ years of experience (doctoral research counts)
• Working from the San Francisco office five days a week, 9-9-5, with at least one interview in person

Bonus points

• RL with verifiable rewards, or learning from sparse or delayed feedback
• Experience with molecular or lab data, uncertainty estimation or active learning
• Publications or open-source work

What's in it for you

• $200k–$300k base + meaningful equity
• A seat on a founding AI team with real data, real labs and real customers
• Visa sponsorship for transfers and renewals (a long process from abroad is harder; please flag your situation early)

Good to know

• Full-time, on-site in San Francisco, 9-9-5
• References carry a lot of weight in this process
• Process: founder conversation → technical interview with the AI lead → research and collaboration interview → reference check → offer

When you apply, please include

• Your CV (PDF), your LinkedIn URL, and Google Scholar or GitHub if you have them
• A training or post-training run you owned from start to finish: what worked, what didn't and why
• Your current location, and confirmation that five days a week in San Francisco works for you
• Your visa or citizenship status
• Your base salary expectation, your notice period, and any other interview processes you're in

Does this sound like you? Apply now. If you're a match, we'll be in touch within 3 working days.

RnB Consultancy | Netherlands | 3604 AB Maarssen | KvK 71489703