Senior Research Engineer - Manipulation

Job description

FieldAI is transforming how robots interact with the real world. Our growing R&D team is based in Boston, where we develop risk-aware, reliable, field-ready AI systems that tackle the hardest problems in robotics and unlock the potential of embodied intelligence. We take a pragmatic approach that goes beyond off-the-shelf, purely data-driven methods or transformer-only architectures, combining cutting-edge research with real-world deployment. Our solutions are already deployed globally, and we continuously improve model performance through rapid iteration driven by real field use.We are seeking a Research Engineer to join our manipulation efforts. In this role, you will work at the intersection of robotics research and applied engineering, building, training, testing, and refining large-scale learned manipulation models and capabilities that accelerate autonomous control and loco-manipulation on humanoid robots. You will collaborate closely with research scientists, engineers, and product partners to design novel manipulation strategies and deliver systems that directly feed into FieldAI’s robot learning pipelines. You will also play a key role in advancing robotics foundation models designed to be generalizable across embodiments, with an initial focus on humanoid platforms. The work will emphasize combinations of learned and physically-grounded models, alongside the data and training systems required to scale them. This role is about pushing the frontier of manipulation research while ensuring that breakthroughs translate into practical, scalable autonomy in real-world environments. FieldAI is transforming how robots interact with the real world. Our growing R&D team is based in Boston, where we develop risk-aware, reliable, field-ready AI systems that tackle the hardest problems in robotics and unlock the potential of embodied intelligence. We take a pragmatic approach that goes beyond off-the-shelf, purely data-driven methods or transformer-only architectures, combining cutting-edge research with real-world deployment. Our solutions are already deployed globally, and we continuously improve model performance through rapid iteration driven by real field use.We are seeking a Research Engineer to join our manipulation efforts. In this role, you will work at the intersection of robotics research and applied engineering, building, training, testing, and refining large-scale learned manipulation models and capabilities that accelerate autonomous control and loco-manipulation on humanoid robots. You will collaborate closely with research scientists, engineers, and product partners to design novel manipulation strategies and deliver systems that directly feed into FieldAI’s robot learning pipelines. You will also play a key role in advancing robotics foundation models designed to be generalizable across embodiments, with an initial focus on humanoid platforms. The work will emphasize combinations of learned and physically-grounded models, alongside the data and training systems required to scale them. This role is about pushing the frontier of manipulation research while ensuring that breakthroughs translate into practical, scalable autonomy in real-world environments. What You’ll Get To Do Advance Humanoid Manipulation Research and Development Design, implement, and evaluate learning-based manipulation models and strategies for humanoid robots across a wide range of tasks. Drive projects from early concepts through on-robot testing and deployment. Develop loco-manipulation capabilities that integrate perception, control, and planning. Drive Robotics Foundation Model Development Contribute to foundation models for manipulation, working on model architecture, data collection, large-scale training pipelines, and deployment infrastructure. Ensure model development supports generalization and transfer across diverse robotic platforms. Collaborate with research scientists to integrate large-scale manipulation data into learning pipelines powering foundation models. Train and evaluate manipulation models using imitation learning, reinforcement learning, and other training approaches. Build Systems That Bridge Research and Deployment Translate research ideas into reliable robotic systems that operate in real-world conditions. Ensure systems are robust, reproducible, and aligned with data collection and learning objectives. Develop experimental infrastructure to support rapid iteration, large-scale training, and evaluation. Collaborate Across Disciplines Partner with mechanical and electrical engineers on hardware integration and system bring-up. Work closely with field teams to refine interfaces and improve manipulation performance. Act as a connective layer between autonomy research and applied robotics engineering. Rapidly Iterate and Deliver Prototype quickly, run experiments on hardware, and validate results in the field. Balance exploratory research with concrete deliverables that support near-term goals. Debug complex system-level issues spanning software, hardware, data, training infrastructure, and learning. What You Have Bachelor’s, Master’s, or PhD in Robotics, Computer Science, Mechanical Engineering, or a related field. 2+ years of hands-on experience in robotic manipulation and/or robot learning in academic or industry settings. Strong foundation in robot kinematics, dynamics, and control, with a focus on manipulation. Strong background in learning-based manipulation, with an emphasis on reinforcement learning and imitation learning, and experience training modern deep-learning models. Experience using frameworks such as PyTorch and building reproducible model-training and evaluation pipelines. Proven experience implementing and evaluating robotic systems on real hardware. Ability to work effectively in fast-paced, highly collaborative environments. Curiosity, ownership, and the confidence to challenge assumptions and propose new approaches. The Extras That Set You Apart Experience wor