About The Position

Warp Sciences brings AI, robotics, and life sciences together in automated laboratories that design, run, and learn from experiments. The company builds infrastructure to help others discover new medicines faster. The Robot Learning Intern will work at the intersection of machine learning, robotics, and real-world experimentation to develop intelligent robotic systems capable of learning complex scientific tasks and performing them reliably in the physical world. The work spans robot learning, simulation, data, and real-world robotics, translating research ideas into working systems.

Requirements

  • Currently pursuing a Bachelor's, Master's, or PhD in Computer Science, Robotics, Machine Learning, or a related field.
  • Strong foundations in machine learning, deep learning, and programming.
  • Experience with PyTorch, JAX, or similar deep learning frameworks.
  • Familiarity with robot learning concepts such as imitation learning, reinforcement learning, or visuomotor policies.
  • Experience with robotics, ML research, simulation, or hands-on technical projects through coursework, research, internships, or personal projects.
  • Strong experimental instincts and interest in understanding why learning systems succeed or fail.
  • Excitement about working with real-world robots and bridging the gap between research and deployment.
  • High ownership, curiosity, and willingness to learn quickly and tackle unfamiliar problems.

Nice To Haves

  • Experience with vision-language-action models, world models, or pretrained robotic foundation models.
  • Experience with reinforcement learning, imitation learning, behavior cloning, or offline RL.
  • Experience with simulation environments such as MuJoCo, Isaac Sim, or Isaac Lab.
  • Experience with sim-to-real transfer, real-to-sim reconstruction, or real-to-sim-to-real learning.
  • Experience learning from human demonstrations, teleoperation, or real-world robot data.
  • Experience with diffusion policies, action chunking, or other modern robot policy architectures.
  • Experience with large-scale multimodal datasets, data pipelines, or distributed training.
  • Experience with dexterous manipulation, contact-rich tasks, or robotic controls.
  • Experience with robot evaluation, benchmarking, and failure analysis.

Responsibilities

  • Develop and experiment with robot learning methods, including imitation learning, reinforcement learning, and post-training of learned policies.
  • Explore learning from human demonstrations, robot experience, and task-specific data to improve robotic manipulation.
  • Work with robot learning architectures such as vision-language-action models, diffusion policies, world models, and other learned representations.
  • Build and improve simulation environments and explore sim-to-real, real-to-sim, and real-to-sim-to-real approaches for training and evaluating robot policies.
  • Help build data pipelines and training workflows for multimodal robotics datasets, including vision, actions, robot state, and other sensor signals.
  • Evaluate learned policies in simulation and on real robots, analyze failures, and iterate on data, models, and training approaches.
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service