ARTIFICIAL INTELLIGENCE / POST-TRAINING INFRASTRUCTURE
Member of Technical Staff, Software Engineer (Backend / AI Infrastructure)
Full-Time • On-Site in San Francisco, CA • $180,000 - $280,000 + Equity
Location
San Francisco, CA
Salary
$180K - $280K
Experience
0-8 Years
Industry
AI / RL Infra
ABOUT THE COMPANY
We are recruiting on behalf of an AI infrastructure company building the post-training and reinforcement-learning pipeline that frontier AI labs rely on. Customers include OpenAI and Amazon AGI Labs. This is not a research project searching for product-market fit. The platform already powers training workflows that some of the most important AI labs in the world depend on daily.
The team is small, flat, and deeply technical (around 10 people), founder-led, and works on-site in San Francisco. Engineers here report directly to the founders and have full autonomy to drive projects end to end.
ABOUT THE ROLE
We are looking for a backend-leaning software engineer to scale and automate the post-training pipeline. You will work in a fast-moving, research-adjacent environment, turning experimental workflows from researchers and co-founders into robust, production-ready systems. You will own backend services and infrastructure end to end, optimizing for scale, cost efficiency, and fast iteration on a primarily Python-based stack.
This is a hands-on, code-every-day role at the core of the RL pipeline. You are building the training loop, not a wrapper on top of someone else's model. This position is on-site in San Francisco, CA. Relocation support is available for the right candidate.
WHAT YOU WILL DO
• Build and optimize automation pipelines that streamline the post-training stack for scale and cost efficiency
• Maintain and support high-concurrency infrastructure that powers customer training pipelines
• Work closely with researchers and co-founders to turn experimental workflows into robust, production-ready systems
• Develop backend services and APIs for environment generation, trace ingestion, and telemetry
• Collaborate on parallelization and coordination of multiple agents across distributed systems
• Ship pragmatic, high-quality software in a flat-structured, deeply technical team of roughly 10 people
WHAT WE ARE LOOKING FOR
Required
• 0-8 years of experience in software engineering, backend, or AI-adjacent work
• Currently working on AI or agent-related projects
• Strong Python proficiency for rapid iteration, and comfort navigating and contributing to large codebases
• Experience with high-concurrency backend infrastructure (FastAPI, queuing, Redis)
• Track record of scaling systems at a startup or big tech, including optimizing for scale and bringing down cost
• Experience building backend systems, data pipelines, or automation infrastructure
• A hands-on engineer who writes code daily (this is not a management track)
• Computer Science degree preferred
• US work authorization (no visa sponsorship available)
Strongly Preferred
• Experience with agent SDKs, LLM tooling, or RL pipelines
• Growth-stage startup experience scaling infrastructure
• Experience coordinating or parallelizing multi-agent systems across distributed systems
• Rust and/or TypeScript experience
• Familiarity with telemetry, trace ingestion, or environment generation
Tech stack: Python, Rust, FastAPI, TypeScript, Agent SDKs, Redis, distributed systems, CPU-based infrastructure.
WHY THIS ROLE STANDS OUT
• You build the training loop, not a wrapper. Engineers here work at the core of the RL pipeline, not on dashboards on top of someone else's model.
• Customers include OpenAI and Amazon AGI Labs. Real traction, not a pitch. You ship software that frontier teams depend on daily.
• Flat structure and founder-led technical culture. You report directly to the founders with full autonomy to drive projects end to end.
• Broad surface area. You will touch backend systems, data pipelines, automation infrastructure, internal tools, and customer-facing prototypes.
COMPENSATION AND DETAILS
• Base salary: $180,000 - $280,000 depending on experience
• Competitive equity
• On-site in San Francisco, CA, with relocation support available for the right candidate
• Full-time, direct hire
• Hiring 2 to 3 engineers for this role
WORK AUTHORIZATION
US work authorization required. Visa sponsorship is not available for this position.
INTERVIEW PROCESS
• Intro Chat (30 minutes): An introductory conversation with a founding team member covering your background, motivations, interest in the company and the problem space, and general fit, along with logistics such as location, relocation willingness, and work authorization. The goal is to assess cultural alignment, communication, and genuine interest.
• Technical Interview (30 to 60 minutes): A collaborative technical interview with a founding engineer or a member of the technical team. The format is an experimental design discussion built around a real problem the team has worked on, for example a parallelization issue coordinating multiple agents. You and the interviewer work through it together, covering baseline experiments, ablations, and experimental design. This is not a traditional coding interview. It evaluates depth of understanding, independent thinking, engagement with the problem, and reasoning ability.
• In-Person Work Trial (1 to 3 days): An on-site work trial at the San Francisco office, working alongside the engineering team on real or representative tasks. This is the primary signal stage, assessing your ability to iterate quickly, work within the codebase, collaborate, and deliver. A minimum of one full day is required, with flexibility for candidates balancing other commitments.