What you need

Strong Python, multi-step LLM agent experience
Deployed/debugged robot manipulation/control policies on hardware
Robotics fundamentals: kinematics, control, ROS 2 diagnostics
Simulation expertise: Isaac Sim, MuJoCo, sim-to-real
GPU service ops: queues, retries, sandboxing, cost controls

What you'll do

Design model-facing tools, APIs, agent loops, skill libraries
Generate/test robot policies in simulation with feedback recovery
Build reproducible evaluations, held-out tasks, regression tests
Serve concurrent agents, isolate code, deploy to robots

Nori Robotics (YC S26) builds affordable humanoids whose intelligence scales with our fleet. We are looking for an engineer who can connect reliable LLM agents with working robot behavior.

Scope: Build agents that turn task descriptions into robot policies, improve them through simulation and execution feedback, and deploy them as a customer-facing service.

What you will work on

• Design model-facing tools and APIs, multi-step agent loops, and reusable skill libraries.

• Generate and test robot policies in simulation, using execution feedback, context management, and retry budgets to recover from failures.

• Build reproducible evaluations, held-out tasks, and regression tests that verify success independently of model self-reports.

• Serve concurrent agents, safely isolate generated code, and deploy tested policies to physical robots while tracking latency, reliability, and compute cost.

Requirements

• Strong Python and experience building LLM agents that run unattended over multiple steps.

• Demonstrated experience deploying and debugging manipulation, imitation-learning, or control policies on physical robots, with measurable results.

• Strong robotics fundamentals: kinematics, coordinate frames, feedback control, calibration, and actuator limits. Ability to diagnose failures across policies, sensors, and hardware using ROS 2 or equivalent.

• Experience with Isaac Sim/Lab, MuJoCo, SAPIEN, or equivalent, including contact dynamics and sim-to-real discrepancies.

• Experience operating GPU-backed services with queues, retries, cancellation, and cost controls; familiarity with sandboxing, network restrictions, and resource limits for generated code.

Nice to have Research in agents, program synthesis, or robot learning; PyTorch and LeRobot; multi-agent orchestration; on-device inference and power profiling; fleet deployment and OTA updates.

San Francisco, CA. Full time. In person.

Compensation. Base pay: $150,000-$200,000 per year, depending on job-related knowledge, skills, and experience.

Optimize your resume for this job

Get a match score and the keywords you're missing

Optimize resume

About Nori

We manufacture Nori: A sub-$2000 bimanual mobile manipulator that can clean rooms, restock shelves, and fold clothes. Nori can handle basic tasks out of the box, and its software lets anyone teach Nori to automate work specific to their home or business. Nori aims to solve the robotics data bottleneck. By deploying affordable robots to customers around the world, we can gather the diverse data required to train the next generation of generalist robotic policies. Every user of Nori makes Nori smarter. Since launching 6 weeks ago we have already deployed our first robot and have over $440k+ in sales. We can bring general-purpose robotics to homes and businesses that could never afford it before. There are 132.7 million households and 8.4 million employer business locations in the U.S. At $1,688 per robot, selling just one Nori to 0.5% of them represents a $1.2B market opportunity in the U.S. alone.

Company profile Website More Software jobs