Member of Technical Staff - Robot Learning

About Warp

At Warp Sciences, our mission is to help the world discover new medicines faster—and build the capacity to do it here in America. We bring AI, robotics, and life sciences together in automated labs that design, run, and learn from experiments, with each experiment making the next one better. We’re not developing drugs of our own. We’re building the infrastructure for the people who do.

Chemistry requires precise, dexterous manipulation in environments filled with instruments, materials, and consumables designed for human hands. Building autonomous laboratories requires robots that can learn the long tail of tasks scientists perform every day, adapt when conditions change, and operate reliably in the physical world.

We’re looking for a Member of Technical Staff, Robot Learning, to build the learning systems that make this possible. You’ll own the loop from expert demonstrations and robot experience through policy training, evaluation, and deployment. You’ll work across robot learning, data, controls, hardware, and chemistry to turn capable models into systems that can perform difficult scientific tasks reliably.

What You’ll Do

• Develop mid- and post-training methods for robot policies using real-world demonstrations, robot experience, and task-specific data.

• Build systems that learn from expert chemists, turning demonstrations of difficult scientific procedures into capabilities robots can reproduce and generalize.

• Develop and adapt world models, world-action models, vision-language-action models, and other learned representations for precise, contact-rich manipulation.

• Build the training and evaluation systems needed to ingest, process, curate, and learn from large-scale multimodal robot datasets.

• Deploy policies on real robotic systems and diagnose failures across models, data, perception, controls, hardware, and task design.

• Close the loop between data collection, training, evaluation, and deployment. Determine what data or training intervention is needed, measure whether it works, and use each deployment to improve the next iteration.

• Partner closely with chemists, hardware engineers, infrastructure engineers, and robot operators to design tasks, collect useful experience, and turn research advances into dependable laboratory capabilities.

What We’re Looking For

We care much more about what you’ve built than your title or number of years of experience.

• Strong foundations in machine learning, deep learning, and robot learning.

• Experience with post-training, imitation learning, reinforcement learning, behavior cloning, or fine-tuning learned policies.

• Experience training, adapting, and debugging modern deep learning models using frameworks such as PyTorch or JAX.

• A track record of designing rigorous experiments and producing measurable improvements in model or policy behavior.

• Strong intuition for data quality and composition: what experience a model needs, where its current data is insufficient, and how to improve it efficiently.

• Experience building data pipelines, training systems, or evaluation infrastructure for multimodal or robotics data.

• Experience working with real-world robotic systems, or a strong track record building learned systems that interact with complex physical environments.

• The ability to investigate failures across the complete system rather than treating the model, data, controls, and hardware as separate problems.

• High ownership and excitement about problems where neither the requirements nor the right technical approach are obvious yet.

Helpful, but Not Required

Experience in any of the following is a plus:

• World models, world-action models, or vision-language-action models

• Post-training or adapting large pretrained models for robotics

• Large-scale multimodal datasets, dataset curation, and distributed training

• Imitation learning, behavior cloning, reinforcement learning, or offline RL

• Learning from human demonstrations, teleoperation data, or real-world robot experience

• Diffusion policies, action chunking, or other modern visuomotor policy architectures

• Dexterous, precise, or contact-rich manipulation

• Multimodal representations spanning vision, language, actions, sensors, and robot state

• Simulation, sim-to-real transfer, or synthetic robot data

• Real-world robot evaluation, benchmarking, and failure analysis

Example Problems You Might Work On

• How do we take a capable pretrained model and post-train it to perform a new chemistry task using demonstrations collected from expert chemists?

• How do we efficiently ingest, curate, and train on millions of trajectories containing video, actions, robot state, language, and other sensor data?

• What data should we collect when a robot can complete 80% of a task but consistently fails during one particular stage?

• How do we determine whether a failure came from the policy, the training data, perception, controls, hardware, or the way the task was specified?

• How do we adapt policies to precise, contact-rich tasks involving powders, liquids, vials, caps, pipettes, and scientific instruments designed for human hands?

• How do we evaluate whether a policy is genuinely becoming more capable, rather than overfitting to a particular setup, object, or demonstration?

What Success Looks Like

You establish a repeatable loop from expert demonstrations and robot experience to training, evaluation, and deployment. Policies become measurably more reliable on real chemistry tasks, and failures lead to clear decisions about what data to collect, what experiments to run, and what parts of the system to improve.

Warp’s robots learn new tasks faster, recover from a broader range of failures, and generalize across materials, equipment, and laboratory conditions. Each round of real-world experience makes the learning system—and the autonomous laboratory built on top of it—more capable.

Equal Employment Opportunity
Warp is an equal opportunity employer. We consider qualified applicants without regard to race, color, religion, national origin, ancestry, sex (including pregnancy, childbirth, breastfeeding, and related medical conditions), gender, gender identity or expression, sexual orientation, age, physical or mental disability, medical condition, genetic information, marital status, military or veteran status, reproductive health decision-making, or any other characteristic protected by applicable law.

Reasonable Accommodations
Warp provides reasonable accommodations to qualified applicants with disabilities in accordance with applicable law. If you need an accommodation during the application or interview process, please contact [email protected]

Fair Chance Hiring
Warp considers qualified applicants with arrest or conviction records in accordance with the California Fair Chance Act and other applicable laws.

General Notice
This job posting describes the position generally and does not constitute an offer of employment or an employment contract.