CAREERS

Responsibilities

As a Senior Software Engineer at Patronus AI, you'll build high-quality simulations used to train, evaluate, and improve AI agents — and the platform tooling that powers them. These environments model real-world workflows and generate the interaction data needed to improve frontier models.

This is a highly autonomous role: you'll own complex agent environments end-to-end, from frontend to backend, and work closely with Applied Researchers to turn environment designs, task specs, and verifier requirements into robust software.

Your work will help frontier labs stress-test and improve the next generation of AI agents, advancing the path toward safe, human-aligned general intelligence.

In this role, you will:

• Build RL environments end-to-end — frontend interfaces, backend services, APIs, data models, and workflows — using technologies like React/TypeScript, Next.js, Python/FastAPI, and Postgres.
• Think critically about task coverage, environment correctness, edge cases, and adversarial agent behavior.
• Work closely with Applied Researchers, subject-matter experts, and QA specialists to ensure the interaction data your environments produce is high quality, and continually find ways to make it better.
• Build tooling and shared services — like agentic systems and frontends for trajectory review, verifier debugging, reward hacking checks, and environment QA — that make every team faster, more consistent, and more confident in the environments they ship.
• Develop reusable patterns and libraries for building simulated worlds across different domains.
• Build the shared infrastructure that lets us run environments across heterogeneous targets (Linux, macOS, mobile, browser-based) and benchmark them against frontier external models and our own for pre- and post-training evaluation.
• Be a power user of AI coding tools (Claude Code, Codex, and similar) — curious about how they work under the hood, with ideas for extending them and building new tooling on top.
• Operate as a senior IC: own architecture, make tradeoffs, drive execution, mentor others, and raise the engineering bar.

Qualifications

“The number one qualification to succeed in this machine learning course is gumption” - John Lafferty, CS Professor at Yale

Above all, we look for a proactive mindset, willingness to learn, relentless drive, and passion for engineering and product. You are a great fit if you have a background in the following:

• 4+ years as a software engineer, full stack engineer, or similar.
• A track record of owning complex systems end-to-end, from ambiguous requirements through production launch — ideally at a startup or on a small, high-velocity team.
• Strong experience building production software with React, TypeScript, and modern frontend patterns.
• Strong experience designing APIs, services, data models, and workflows — ideally with TypeScript, Go, Next.js, Python, FastAPI, and Postgres.
• A solid understanding of how modern LLMs work at the API level — tool calling, agent loops, context management — and experience building custom tools, harnesses, or non-trivial agent integrations.
• Strong intuition for edge cases, correctness, failure modes, and adversarial behavior in agentic systems.
• Familiarity with Kubernetes, distributed systems, sandboxing, orchestration, or platform engineering.
• Daily user of agentic coding tools (Claude Code, Codex, Cursor, or similar) with instincts for where to build automations, extensions, or entirely new tools on top.

Nice to have:

• Experience with reinforcement learning, agent evaluation, verifiers, reward models, or ML infrastructure — though deep applied research experience isn't required.
• Experience with Playwright, Selenium, or browser automation.
• BS, MS, or PhD in Computer Science, Machine Learning, Statistics, Mathematics, or a related quantitative field — or equivalent experience.