Machine Learning Engineer New Grad
Preference Model
Machine Learning Engineer New Grad
Preference Model
Reinforcement learning environments for frontier AI
Preference Model is a superintelligence data research company. We build RL environments for training capable, better-aligned superintelligences.
Almost every aspect of how a model behaves is shaped by the reward signals it's trained on. The biggest problem with AIs today is that they don't always do what we intend. Sometimes that's because they aren't capable enough, and sometimes it's because they aren't aligned. Both problems come down to the training objectives.
We're trying to address both of these issues in the most direct, highest leverage way available to us, which today means researching how to build RL environments better, not just for the AIs of today but for the AIs as they will be when they outsmart us.
Over the past year, we've built RL environments for several frontier labs, and we're backed by $16M in seed funding led by a16z. Our founding team has previous experience on Anthropic’s data team building data infrastructure, and datasets behind Claude.
We’re hiring new graduate Machine Learning Engineers to design and build reinforcement learning environments to safely advance model capabilities in machine learning research and engineering. Specifically, you'll be teaching frontier models to do the work of an ML engineer or researcher at a frontier lab.
This role blends research and engineering. It will require you to stay up to date with the latest research, develop novel approaches, and realize them in code. You will have full ownership and autonomy of the environments you build. Your work will include designing and implementing RL environments, conducting experiments and evaluations, delivering your work into production training runs, and collaborating with other researchers and engineers.
You will join our Capabilities org, a small, high-ownership team and contribute directly to the data layer that powers frontier LLM capability.
Note: this role is for recent graduates only who can start soon. We are prioritizing PhDs who have research publications and experience
Design and build RL environments and reward schemes that produce clean, learnable signals for frontier models on ML research and engineering tasks.
Build deep expertise across the frontier of ML research, training, and inference infrastructure.
Collaborate with others to brainstorm and create new ideas and tools to improve the environment building process.
You have strong ML fundamentals and broad research interests. You read many papers or tutorials, understand topics deeply and have the creativity to translate them into RLVR problems.
Expert knowledge in an active DL/ML research area, with publications or public code to show for it.
Research experience (PhD) is a strongly preferred.
Deep understanding of transformer internals
Proficiency in Python, Numpy, and systems programming; ideally PyTorch or JAX
Smart problem solvers who take ownership and drives solutions end-to-end
Passion for staying current with the rapidly evolving ML infrastructure landscape
Ability to meet throughput expectations and respond quickly to feedback
Strong expertise in kernel development (CUDA, Triton, Pallas), optimizing non-trivial neural modules to specific hardware
Research projects, coursework, or personal work involving RL environments (any framework, any scale)
Open-source contributions to ML infrastructure or RL tooling
Experience with any cloud platform (AWS, GCP, Azure) or infrastructure-as-code tools
Competitive cash and equity compensation (>90th percentile)
Ownership and autonomy in a fast moving startup environment
Opportunity to work with top machine learning engineers
Health, vision, dental, benefits
401K match
Lunch provided everyday onsite
Weekly snack orders
Visa sponsorship & relocation support available
We value diverse perspectives and experiences. If you're excited about this role but don't check every box, we still encourage you to apply.
Reinforcement learning environments for frontier AI
Preference Model builds reinforcement learning environments and harnesses to train frontier AI models, focusing on how reward signals shape model behavior. It designs robust RL harnesses and simulators that treat reward functions and graders as an attack surface, testing them across millions of training rollouts to harden against loopholes and gaming, with fail-safes for safety. Unlike others that only build models or tools, it targets the reward design and alignment bottlenecks, drawing talent from Anthropic, Stripe, Google DeepMind, and others, and backed by notable investors. Its goal is to broaden access to capable AI while preserving strong safeguards to keep frontier systems safe.
Explore more opportunities at Preference Model