Senior Robotics Manipulation Engineer (Talent Network)

Dexterity; Redwood City (CA); Sur place; USD 150 000 - 210 000; Plein temps

Résumé du poste

Dexterity is seeking a Senior Robotics Manipulation Engineer in Redwood City, CA to advance multi-sensor robotic systems through data-driven control and learning-based methods.

You will design, train and deploy learned policies for manipulation, force control and motion planning, collaborating across ML, robotics, and product teams to deliver scalable software and robust robotic solutions.

Qualifications

• Experience in robotics or physical AI including manipulation, control and learning.
• Strong software engineering practices for maintainable, testable code.
• Experience with reinforcement learning, imitation learning and modern ML methods.

Responsabilités

• Develop data-driven robot control systems and learned models for manipulation and force control.
• Conduct sensor fusion and state estimation to leverage multimodal data.
• Collaborate with ML engineers, hardware and product teams to deliver robust solutions.
• Build and maintain high-quality, scalable software for robotic platforms.
• Train, evaluate and improve models using large-scale data from real robots.
• Lead initiatives to improve model performance and rigor across the stack.

Connaissances

Python programming

C++

Linux

Reinforcement learning

Task & Motion Planning

Distributed ML training

Git & CI pipelines

Docker & Kubernetes

Formation

BS/MS/PhD in CS or related

Outils

Ray

Linux

Docker

Kubernetes

Git

Description du poste

• As a Senior Robotics Manipulation Engineer, you will be working on a myriad of challenges related to manipulation, control, simulation and learning
• You will bring a data driven approach to learned robot controls
• You will also stay abreast of the latest progress in reinforcement learning, motion planning and controls, and other related fields in order to further develop Dexterity’s technology foundations in physical AI
• Dexterity’s robotic solutions integrate data from a multitude of sensors, including RGB cameras, depth sensors, force-torque sensors, encoders, system telemetry and human input
• To better inform the planning and execution system, you may work on sensor fusion and state estimation techniques to leverage this multimodal sensory data
• You will work closely with the data platform, simulation, and robot operations teams to collect training data to improve your models
• In addition to core manipulation and controls, you will be responsible for building high quality, maintainable, and performant software
• Train models for robotic manipulation, force control, servoing etc
• Convert heuristics based control to learned models
• Develop the right state space representation for our Robotics Systems
• Collaborate with roboticist, machine learning engineers, mechanical engineers, and product managers to find optimal system solutions
• Implement, test, and maintain high quality software
• Lead efforts to improve the rigor and performance of models within Dexterity

Ideally, you will possess strong software system and interface design skillsExperience with multiple of the following:BS/MS/PhD in Computer Science or a related disciplineDistributed ML Training Platforms (Ray or other)Strong Python programming skills5+ years of relevant Industry or Research experienceTask and Motion PlanningKnowledge of Modern C++ and LinuxArtificial Intelligence (imitation learning, reinforcement learning, deep learning, LLMs)SimulationExperience with Git and modern CI pipelinesPrevious startup experienceExperience with Docker and Kubernetes