Member of Technical Staff — Agentic RL Infrastructure
- San Francisco · CA
- Publicada hace 56 días
Verificada en la fuente hace 2 h
Con tu cuenta gratuita de Joonow, vas al sitio de la empresa y la vacante queda en tus postulaciones.
Ninguna empresa seria cobra cuotas, cursos, exámenes ni uniformes. ¿Te pareció raro?
Descripción del empleo
Introducing Moonlake, AI for creating world simulations.
About Moonlake
Moonlake is building the frontier of interactive world models: systems that generate, simulate, and reason over 3D environments for embodied AI, robotics and gaming. We develop the simulation infrastructure to build worlds (e.g., assets, scenes, digital twins) at scale.
Our team sits at the intersection of
Embodied AI
Robotics simulation
Interactive 3D worlds
World models
Real-time generation
AI infrastructure
Moonlake is building the next generation of AI infrastructure for interactive digital worlds. Our mission is to enable anyone to create, simulate, and interact with rich environments using natural language and multimodal inputs, turning simple ideas into worlds with structure, logic, and agents that can perceive and act.
Our team has raised $28M in seed funding from NVIDIA Ventures, Threshold Ventures, AIX Ventures and notable angels including Naval Ravikant and Jeff Dean to build the foundational layer for the future of AI - powering everything from robotics training, simulations, and digital twins. Our goal is to make building and experimenting with these environments as accessible and scalable as publishing video on the internet.
We are looking for exceptional research engineers and applied researchers to help push the frontier of interactive AI.
The Role
We're looking for an exceptional Member of Technical Staff — Agentic RL Infrastructure to build the foundational infrastructure for training the next generation of intelligent agents in simulated environments. This is a rare opportunity to work at the intersection of robotics, reinforcement learning, distributed systems, and Physical AI.
You'll work closely with researchers and engineers to design, build, and scale the infrastructure that enables vision-language models (VLMs) and autonomous agents to learn through interaction with physically realistic environments. From developing simulation pipelines to orchestrating thousands of parallel training environments, you'll own critical systems that make large-scale robotics reinforcement learning possible.
This is a hands-on engineering role for someone who thrives in early-stage startups, enjoys building complex systems from the ground up, and is excited to apply strong software engineering fundamentals to frontier AI research.
What You'll Do
Design and build scalable infrastructure for robotics reinforcement learning and vision-language model post-training.
Develop simulation environments and training pipelines using frameworks such as MuJoCo, NVIDIA Isaac Sim, Isaac Lab, and PhysX.
Architect systems capable of running thousands of physics-based simulations in parallel to support large-scale agent training.
Build end-to-end infrastructure spanning simulation orchestration, distributed compute, data pipelines, experiment management, and developer tooling.
Optimize simulation throughput, resource utilization, and training efficiency across GPU and CPU infrastructure.
Collaborate closely with AI researchers to translate experimental training requirements into reliable, scalable engineering systems.
Benchmark simulation frameworks, identify performance bottlenecks, and make architectural trade-offs between simulation fidelity, scalability, and cost.
Develop tools and workflows that enable researchers to rapidly create, test, and iterate on new robotics tasks and training environments.
Own infrastructure projects end-to-end, from initial architecture and prototyping to production deployment and optimization.
What We're Looking For
Strong full-stack software engineering experience with the ability to design, build, and deploy complex systems end-to-end.
Experience working at an early-stage startup or building technically ambitious products from zero to one.
Strong proficiency in Python and experience building backend infrastructure, APIs, distributed systems, or cloud-native applications.
Some hands-on experience with robotics, reinforcement learning, simulation environments, or machine learning infrastructure.
Familiarity with robotics simulation frameworks such as MuJoCo, NVIDIA Isaac Sim, Isaac Lab, or PhysX.
Understanding of distributed computing, parallel processing, GPU infrastructure, and performance optimization.
Ability to reason about system architecture, identify bottlenecks, and make pragmatic engineering trade-offs.
Comfortable working across infrastructure, backend systems, developer tooling, and research engineering.
Ability to operate independently in ambiguous environments, rapidly prototype solutions, and take ownership of technically challenging problems.
Experience with large-scale RL training, VLMs, embodied AI, or robotics simulation infrastructure is a plus.
Why This Matters
The next frontier of AI is moving beyond language and into the physical world. For intelligent agents to reason, plan, and act in real environments, they must first learn through interaction with simulations that are realistic, diverse, and scalable. Building that infrastructure is one of the fundamental challenges in Physical AI.
At Moonlake, you'll build the systems that make large-scale robotics training possible, enabling researchers to train and evaluate intelligent agents across thousands of simulated environments. Your work won't just accelerate individual experiments; it will help establish the infrastructure foundation for a future where intelligent machines can learn to operate in the physical world.
Clarify the expected seniority levelMake the role less repetitive
We are committed to being an on-site, in-person team currently based in San Francisco.