Senior On-Policy RL Engineer — Robotics Autonomy

reflexrobotics; New York (NY); On-site; USD 140,000 - 210,000; Full time

Job summary

Reflex Robotics in New York designs AI-powered humanoid robots for tough industrial tasks. We are a small, mission-driven startup backed by top investors and real hardware with profitable unit economics.

We’re seeking an on-policy RL engineer to build robust robot policies, debug gradients, and push performance toward near-perfect deployment. You’ll own meaningful contributions and impact from day one, shaping the product’s future across real-world use cases.

Qualifications

• You re-implemented core RL algorithms (SAC, DDPG) and can debug unstable gradients.
• You contributed to sample-efficient RL methods (DreamerV3, MuZero).
• You shipped on-policy RL on real hardware (e.g., robotics deployments).

Responsibilities

• Develop robust on-policy RL policies for robotic systems.
• Debug gradients and tune hyperparameters for stable learning.
• Ship RL solutions on hardware in real-world environments.

Skills

RL reimplementation (SAC, DDPG)

DreamerV3 / MuZero contributions

On-policy RL deployment on hardware

Job description

Reflex Robotics in New York designs AI-powered humanoid robots for tough industrial tasks. We are a small, mission-driven startup backed by top investors and real hardware with profitable unit economics.

We’re seeking an on-policy RL engineer to build robust robot policies, debug gradients, and push performance toward near-perfect deployment. You’ll own meaningful contributions and impact from day one, shaping the product’s future across real-world use cases.