About The Position

Trener is hiring a Robotics Research Engineer (VLA Engineer) for T-Labs to turn vision-language-action research into reliable robot behavior for production-line use. The role owns the full development loop for foundation-model-based manipulation policies, including data collection and preparation, model training, evaluation, and deployment on physical robots. The engineer will work with AI compute, perception, and integration engineers to advance models from research to pilots and customer deployments, while addressing the variability and reliability demands of industrial work cells. The position focuses on industrial manipulation tasks such as pick-and-place, kitting, machine tending, and assembly. It includes building robot-learning data pipelines, adapting pretrained vision-language models to robot action spaces, establishing real-world evaluation benchmarks, closing the sim-to-real gap, and deploying policies through ROS 2 to edge hardware. The engineer will also track and reproduce research, communicate findings internally and potentially through publications, and collaborate with product, integration, and customer-facing teams to translate customer needs into measurable tasks and milestones.

Requirements

  • MS or PhD in Robotics, Computer Science, Electrical or Mechanical Engineering, or a related field, or equivalent industry experience.
  • At least 3 years of hands-on experience in robot learning, imitation learning, or reinforcement learning applied to manipulation.
  • Direct experience training or fine-tuning large multimodal models, such as VLMs, VLAs, or transformer-based policies, in PyTorch or JAX.
  • Strong Python skills and working proficiency in C++.
  • Experience running learned policies on physical robot hardware, not only in simulation.
  • Familiarity with ROS or ROS 2.
  • Comfort with distributed GPU training and experiment tracking, such as Weights & Biases or MLflow.
  • A track record of shipping research into working systems and communicating clearly across research and engineering teams.

Nice To Haves

  • Publications on robot learning or embodied AI at CoRL, RSS, ICRA, IROS, NeurIPS, ICML, or CVPR.
  • Contributions to open-source robot-learning projects.
  • Experience with diffusion policies, flow matching, or world models for control.
  • Hands-on experience with industrial arms such as UR, ABB, KUKA, or FANUC, and with grippers, force/torque sensing, or tactile sensing.
  • Experience optimizing models for edge inference using quantization, pruning, distillation, TensorRT, or NVIDIA Jetson.
  • Background in large-scale teleoperation or data-collection infrastructure.
  • Experience at a robotics startup or in a production manufacturing environment.

Responsibilities

  • Design, train, and fine-tune Vision-Language-Action models for industrial manipulation tasks, including pick-and-place, kitting, machine tending, and assembly.
  • Build and scale robot-learning data pipelines for teleoperation and demonstration collection, data curation, augmentation, and synthetic data generation from simulation.
  • Adapt pretrained vision-language backbones to robot action spaces, including action tokenization, diffusion or flow-matching action heads, and multi-camera inputs.
  • Define evaluation protocols and benchmarks for real-world success rate, robustness, and generalization to new parts, cells, and instructions.
  • Use simulation environments and domain randomization to close the sim-to-real gap.
  • Deploy policies on physical robots through ROS 2 and work with the AI Compute team to meet edge-hardware latency and memory targets using techniques such as quantization, distillation, and TensorRT.
  • Track and reproduce state-of-the-art research and share findings through internal tech talks, write-ups, and, where appropriate, publications.
  • Partner with Product, Integration, and customer-facing teams to translate customer use cases into learnable tasks and measurable milestones.
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service