Robotics Researcher, Teleoperation

About the company

Efference provides high-performance perception and compute infrastructure for training and deploying robot models in the cloud. Our vision is to make robotic deployments scalable with shared compute.

We approach the problem by owning the entire path from the image sensor to the datacenter GPU. We shrink the load at the sensor, fix the uplink so robot traffic is fast and reliable, and clock synchronize the system so observations arrive time-aligned, ready for inference.

About the role

You'll develop teleoperation systems for collecting useful robot demonstrations and operating robots through Efference's perception and network stack. You'll study how video quality, viewpoint, delay, and control feedback affect operator performance, then use those results to improve the product.

What you’ll do

• Build teleoperation workflows using leader-follower devices, VR interfaces, or other input systems suited to the robot and task.
• Develop retargeting, calibration, and command interfaces with controls engineers, accounting for kinematic differences and the limits of each platform.
• Integrate synchronized video, robot state, and operator input, with instrumentation for the delay from observation to display and command to execution.
• Run experiments on bitrate, frame rate, depth cues, camera placement, jitter, and dropped observations to measure task quality and operator workload.
• Record demonstrations with aligned actions and observations, preserving session metadata, interventions, and quality flags for training and evaluation.
• Implement and test pause, handoff, reconnect, and local fallback behavior with the controls team; use operator feedback to improve difficult workflows.

What we are looking for

• Experience building teleoperation, robot learning data collection, or interactive robotics systems on physical hardware.
• Strong Python or C++ skills and an understanding of robot kinematics, coordinate frames, calibration, and control interfaces.
• Practical knowledge of real-time video, network delay, and the synchronization of heterogeneous sensor and action streams.
• Ability to design experiments with operators and distinguish an interface problem from a control, perception, or network problem.
• Experience producing documented, usable demonstration data and maintaining the equipment and software needed to collect it.

Nice to Have

• Experience with OpenXR, Unity, stereo rendering, hand tracking, haptics, or wearable data collection.
• Work on bimanual manipulation, humanoid retargeting, shared autonomy, or learning from demonstrations.
• Experience with remote operation over variable networks and evaluating operator comfort and performance over long sessions.