Robot Learning Generalist

Generalist

About The Position

About the Role: This is a deeply cross-functional, execution-oriented role at the center of our robotics stack. As an Robot Learning Generalist, you will ensure that robots, data, models, and evaluations all come together into a tight, high-velocity feedback loop. You won’t sit purely in ML research, robotics controls, hardware, or operations — you will operate horizontally across all of them. In practice, you will help design tasks, coordinate robot data collection, kick off training jobs, run evaluations, analyze results, and ensure robots are physically ready for rollouts. You may build physical benchmarks, lightly modify hardware setups, test third-party tooling, and write documentation that enables others to replicate and scale your work. This role is about making the entire system move faster. If something is blocking model progress — whether it’s missing data, unclear task definitions, misconfigured hardware, broken eval scripts, or unclear documentation — you fix it, or find the right person to fix it. You are the connective tissue between research, infrastructure, hardware, and field operations. You might thrive in this role if you: Have hands-on experience with robot data collection, evaluation, or deployment Have trained or fine-tuned ML models and understand the full lifecycle from data → training → evaluation Are comfortable running experiments and tracking real-world metrics across multiple model variants Enjoy operating across software, hardware, and physical systems Have some exposure to basic EE/ME tasks (wiring, mounting sensors, assembling fixtures, debugging hardware) Are highly organized and can coordinate multiple moving parts simultaneously Write clear, structured documentation Prefer execution and iteration speed over theoretical purity Like being the person who “just makes it work” What This Role Is Not You will be a part of the ML team, and working very closely with ML, brainstorming ideas, and may prototype as well. However: This is not a pure ML research role focused on designing new model architectures or advancing core learning algorithms. This is not a large-scale infrastructure engineering role building distributed systems, databases, or UI platforms. This is not a deep robotics controls or firmware engineering role. Instead, this role sits at the intersection of ML, robotics, and operations — ensuring our systems run end-to-end in the real world, and improving them through tight execution loops. If you are most excited by hands-on iteration, cross-functional execution, and accelerating the entire system — this role may be a strong fit. Growth & Impact This role sits on the front lines of scaling our robot fleet — both for internal R&D and external product deployments. As our robots expand in number, capability, and customer exposure, the complexity and leverage of this role grows with it. Career growth here is directly tied to the scale and impact of our robots in the real world. The more surface area you can reliably operationalize — across data collection, training loops, evaluation rigor, and deployment readiness — the more influence and ownership you will earn. If you want to grow alongside a rapidly scaling embodied AI system and shape how robots move from research to widespread deployment and applications, this role offers unusually high leverage.

Requirements

Have hands-on experience with robot data collection, evaluation, or deployment
Have trained or fine-tuned ML models and understand the full lifecycle from data → training → evaluation
Are comfortable running experiments and tracking real-world metrics across multiple model variants
Enjoy operating across software, hardware, and physical systems
Have some exposure to basic EE/ME tasks (wiring, mounting sensors, assembling fixtures, debugging hardware)
Are highly organized and can coordinate multiple moving parts simultaneously
Write clear, structured documentation
Prefer execution and iteration speed over theoretical purity
Like being the person who “just makes it work”

Responsibilities

Designing new robotic tasks and benchmarks to evaluate model capabilities
Procuring materials and building lightweight physical benchmarks (e.g., Montessori task boards, manipulation setups)
Coordinating robot data collection efforts across internal operators and partners
Ensuring robots are properly configured, calibrated, and ready for rollouts and evaluations
Launching model training jobs and tracking experiments across multiple variants
Running structured evaluations and measuring real-world success rates
Analyzing results and closing feedback loops with ML researchers
Beta testing internal and third-party tools for teaching robots new skills
Writing clear documentation and playbooks so others can reproduce workflows
Identifying operational bottlenecks and improving system throughput end-to-end