Reinforcement Learning Robotics Jobs in Kentucky

Location: San Francisco Bay Area
Employment Type: Full-Time
Level: Executive / Technical Leadership
Work Arrangement: On-Site / Highly Collaborative

An early-stage robotics company is building a new generation of humanoid robotic systems designed to allow skilled workers to perform complex physical tasks remotely and safely.

The company's approach combines real-time teleoperation, haptic interfaces, physical-world data collection, and robot learning. Rather than treating teleoperation as an end state, the platform is designed to create a continuous feedback loop: skilled operators use the robots to perform real-world work, those interactions generate rich training data, and that data becomes the foundation for increasingly capable robotic autonomy.

We are seeking a VP, Robot Learning to build and lead the company's robot learning function and define the technical roadmap from teleoperation data through learned policies and eventual autonomous operation.

This is not a research-only leadership position. The successful candidate will need to connect research with a live robotic product, make pragmatic technical decisions, and build systems that work reliably in the physical world.

• Define and own the company's end-to-end robot learning roadmap, from data collection and structuring through policy learning and deployment on physical robots.


• Establish the technical strategy for converting teleoperation and physical-world interactions into high-quality training data.


• Develop approaches for learning from real-world demonstrations, including contact-rich interactions, force feedback, failures, recovery behaviors, and other information that may not be captured by conventional visual datasets.


• Translate the company's broader product and autonomy vision into a practical technical roadmap.


• Determine the appropriate combination of imitation learning, reinforcement learning, perception, control, and other learning approaches needed to advance robotic capabilities.


• Establish the architecture and processes required to move from raw teleoperation data to usable training datasets and deployed policies.


• Work closely with hardware, controls, and software engineers to ensure learning systems are designed around the capabilities and limitations of the physical robot.


• Establish research and engineering priorities while balancing technical ambition with product and deployment requirements.


• Build, mentor, and eventually lead the robot learning organization as the company scales.


• Establish technical standards, development processes, and a culture of rigorous experimentation and real-world validation.


• Serve as a senior technical leader and thought partner to the executive team on the future of robotic autonomy.

• PhD or equivalent depth in robotics, computer science, machine learning, or a closely related technical field.


• Significant hands-on experience in robot learning, manipulation, embodied AI, or related robotics research and development.


• Demonstrated experience taking robotics learning systems beyond research environments and into physical-world applications.


• Strong understanding of areas such as:
  
  
  
  • Imitation learning
  
  
  • Robot learning
  
  
  • Manipulation
  
  
  • Motion and control
  
  
  • Perception
  
  
  • Reinforcement learning
  
  
  • Demonstration-based learning
  
  
  
  


• Experience working with data generated from physical robotic systems, teleoperation, human demonstrations, or other real-world interaction.


• Strong technical judgment and the ability to distinguish promising research directions from approaches that can realistically become production systems.


• Experience defining technical roadmaps and operating with substantial autonomy.


• Demonstrated ability to build or lead a technical organization, team, or major research/product function.


• Strong communication skills and the ability to work across research, software, hardware, controls, and product disciplines.


• A builder mentality and willingness to operate hands-on in an early-stage environment.

Candidates may come from advanced robotics research organizations, robotics startups, university labs, or other teams working at the intersection of machine learning and physical systems.

Especially relevant backgrounds include:

• Robot learning for physical manipulation.


• Learning from teleoperation or human demonstrations.


• Haptic or force-feedback-based robotics.


• Contact-rich manipulation.


• Learning from real-world failure and recovery behaviors.


• Developing learning pipelines connected to commercial robotic products.


• Taking robotics research from prototype through physical deployment.


• Building a robotics or machine learning organization from an early stage.

Experience exclusively focused on simulation, benchmarks, or offline datasets without meaningful physical-robot deployment is less directly aligned with this opportunity.

This is an opportunity to establish the robot learning function at a company building from a relatively clean slate.

You will have substantial influence over the technical architecture, research priorities, team structure, and path from today's teleoperated systems toward increasingly capable autonomous robots.

The role is particularly suited to someone who wants founder-level technical ownership without necessarily being the founder—someone who wants to define the roadmap, build the team, and have a direct impact on how the company's robotics platform evolves.

The company is currently operating at an early stage, with compensation structured around a combination of cash compensation and meaningful equity ownership.

The opportunity is best suited to candidates who are motivated by significant scope, technical influence, and long-term equity upside and who are comfortable joining before the company reaches its next stage of funding and scale.