Want to know if this job is worth applying to?
San Francisco, California, United States
Onsite
Our mission is to automate coding. The first step in our journey is to build the best tool for professional programmers, using a combination of inventive research, design, and engineering. Our organization is very flat, and our team is small and talent dense. We particularly like people who are truth-seeking, passionate, and creative. We enjoy spirited debate, crazy ideas, and shipping code.
One of our north stars is a future where teams can hire Grok as a remote colleague. Reaching that future requires training our models in realistic, diverse environments that support complex, end-to-end work across domains.
As a Software Engineer on the RL Environments team at SpaceXAI, you’ll build the systems that turn real-world data into high-quality environments and tasks for reinforcement learning. You’ll create dramatically more realistic training environments while owning the shared platforms and quality layers that help us produce trustworthy data quickly and at scale.
This role sits at the intersection of research, data, and engineering. You’ll work closely with ML Platform and research teams to turn company data, Grok Bot interactions, vendor-built tasks, tutor data, acquired data, and synthetic data into training-ready environments. Your work will make high-quality RL data easier to create, validate, discover, and use across our model-training efforts.
Building platforms that allows us to build complex, realistic, and diverse RL environments at scale, that support end-to-end tasks across knowledge-work domains.
Designing end-to-end factory that turns massive raw data into useful and realistic environments.
Defining and applying consistent quality standards across vendor, tutor, acquired, and synthetic data.
Creating self-serve APIs and tooling that accelerate task and environment development for internal teams and external contributors.
Partnering across research, platform, and external teams to translate model-capability goals into effective training tasks and environments.
You have strong software engineering fundamentals and experience with data platforms, developer tools, distributed systems, or ML infrastructure.
You can turn ambiguous quality standards into concrete, automated checks.
You’re comfortable building repeatable pipelines from messy, heterogeneous data.
You care about model behavior and can translate capability goals into tasks and experiments.
You collaborate well across disciplines and own open-ended problems end to end.
If there appears to be a fit, we'll reach out to schedule 2-3 short technicals. After, we'll schedule an onsite in our office, where you'll work on a small project, discuss ideas, and meet the team.
