About the role

Help engineers fix what matters, automatically. You’ll help build and improve our AutoFix system: AI-powered agents that generate infrastructure code fixes at scale. You’ll work at the intersection of static analysis, developer tools, and applied AI. You’ll ship fast, collaborating deeply with customers, and solving hard problems that sit at the core of our product.

What we’re looking for:

• GMT+2 to GMT-6 time zone

• You’ve built production-grade AI agents: systems that reason through tasks, call tools or APIs, validate outputs, and handle real-world edge cases like retries, failures, and observability. Bonus if you’ve used tools like LangChain, Hugging Face, Qdrant/pgvector and implemented concepts like RAG, agent workflows or model context protocols (MCP).

• Strong Golang (or equivalent) experience. You write idiomatic, concurrent code, can track down memory leaks, and debug race conditions.

• Familiarity with Terraform. You’ve written Terraform before and understand how remote modules work, and can make sense of a wide variety of project structures including the messy, inconsistent ones you find in the real world.

• You move fast. You get a dopamine hit every time you release to production, and jump on issues without hesitation.

• You've built something from scratch that you're proud of; ideally in infra/devtools or AI-powered automation.

• You’ll thrive in an amazing, experienced, hardworking, respectful, supportive, and fun team.

• (Preferred) Familiarity with CloudFormation and Kubernetes.

Examples of challenges we have worked on recently:

• Automatically fixing infrastructure issues. With infrastructure changes there's a lower tolerance for AI-generated slop - there's not the same safety nets in terms of testing and the risk is often higher. We've been iterating on our system that combines the AI-generated changes with our best-in-class static analysis engine to robustly open good-quality PRs to fix the most important issues for our customers.

• Built the best-in-class Terraform static analysis engine. We needed a way we could parse massive Terraform mono-repos as quickly as possible, able to handle tens of thousands of module downloads, as well as gracefully handling incomplete data.

• Scaled to support customers with thousands of GitHub organizations and tens of thousands of repositories. We overhauled our APIs, interfaces, onboarding processes, infrastructure, and more.

What we value:

1. Customer, not customer: It is all about seeing us and the customer as one. We like to be a part of the user’s team, and help them however we can. If the user is not successful, then we will not be either so we try to walk in their shoes. It\'s more than work - we build relationships and community with users and customers.

2. Open is our core: Put yourself out there. Show your learning. Transparency builds confidence. Encourage sharing the good and the bad. The best decisions are made when everyone has access to all the data. Be straightforward and kind, feedback is about your work not your person.

3. Let\’s JEDI: Let’s Just Effing Do It! Own it and move fast. A good plan fiercely executed now is better than a perfect plan later. We ask for help and unblock each other. The main thing is to keep the main thing the main thing.

Benefits:

• Fully remote team

• Two meetups a year - last year we went to Croatia and Barcelona.

• Employee-friendly equity terms, including a 10 year exercise window

• 401k matching (US)

• Health, dental, and vision insurance (US)

• 31 days paid leave per year (includes national holidays)

• 12 weeks paid parental leave

About the interview

• 25 minute initial chat

• Two or three 55 minute technical interviews (including code walkthrough)

• 55 minute value-based interview

About Infracost

We\u2019re changing the way cloud costs are managed. Infracost helps engineers shift FinOps left by enabling proactive cost issue detection and automated fixes. We plug into developer workflows (like GitHub and Azure Repos), show cost impact in pull requests, enforce tagging and FinOps best practices, and generate PRs to fix issues automatically.

We\u2019re backed by Sequoia, YC and trusted by Fortune 500 enterprises. You\u2019ll join a small, experienced, and supportive team that\u2019s shipping fast, solving real infrastructure problems, and having fun while doing it.

Whether you\u2019re an engineer tackling complex systems, a product manager shaping strategy from real customer pain points, or a customer success lead working directly with users; there\u2019s meaningful work here for you. If you care about cloud efficiency, great UX, and helping teams move faster and smarter, we’d love to work with you!