> Markdown version of [/jobs/ext/2713227-engineer-systems-infrastructure](https://www.wearedevelopers.com/jobs/ext/2713227-engineer-systems-infrastructure). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md). --- # Engineer, Systems (Infrastructure) - **Company:** Ai, Inc - **Location:** San Francisco, CA, United States - **Experience:** Experienced - **Salary:** $200,000.0 - $300,000.0 - **Contract:** Permanent contract - **Skills:** Artificial Intelligence, Databases, Linux, File Systems, Github, PostgreSQL, Redis, TypeScript, Google Cloud, Large Language Models, Kustomize Configuration Management, Kubernetes - **Published:** September 4, 2026 - **Apply:** https://startup.jobs/engineer-systems-infrastructure-adapt-company-9785745 ## About the Role Years aren't all equal (startup years are like dog years, as an example) and they're not the best predictor of success. That being said, you probably exhibit most of these characteristics: * 4-7+ years of professional engineering experience, with a meaningful chunk of it on infrastructure * Experience working at an early stage (seed, Series A, Series B) startup, ideally one where the infrastructure was the product. A build platform, CI system, hosting provider, orchestrator, or database * You've owned a multi-tenant compute or build system where users ran arbitrary code on your machines and you were accountable for isolation, reliability, and cost * You're the person teammates bring an architecture decision to before they commit to it, and you've mentored engineers without holding a manager title * Strong in at least one systems language (e.g. Rust or Go). We use Rust for performance-oriented services, and Typescript to glue them together. * Comfortable at the Linux layer: cgroups, namespaces, filesystems, networking. You've read kernel documentation on purpose * Experience with one major cloud provider (we're on GCP) and Kubernetes as a tool you use directly, not one you hand to a platform team * Deep familiarity with container and microVM runtimes like gvisor and Firecracker. * Working fluency in full-stack TypeScript, or the appetite to get there quickly. You will ship in the product too * Strong interest in many of the latest and greatest AI tools, models, and infrastructure market evolution * You've carried a pager (maybe not literally) for something people depended on, and the system got better because you did * Ability to break down a complex and large problem into independently shippable solutions (you may have found yourself saying "What's the simplest thing we can do to try to solve this problem?") * Care deeply about users: infra work here is judged by how great a customer's experience is with the product. * A genuine desire to grow the craft of engineering and product-building (you could call it "taste"), with an eye for the little big details in both architecture and product ## Description We're hiring an engineer to own the computer underneath Adapt. This is infrastructure in service of the product, not infrastructure for its own sake. We are hiring for a systems expert, but all engineers at Adapt are expected to be in service ultimately of the product we're building and the customers that use it. That means that your focus will always be on world-class systems that power customer workloads. This role's center of gravity is the layer underneath, but it is not a platform team walled off from the product. You'll sit in customer conversations, ship user-facing features, and follow your work all the way out to the people using it. The systems in play: * Sandbox: Firecracker-based microVM orchestration with sub-second boots for agents. * Storage: Distributed filesystem backed by object storage, allowing concurrent read/write from multiple sandboxes. * Infra: Google Cloud Platform, GKE, Kustomize, GitHub Actions, Postgres, Redis, and the tracing and metrics for all of it. Today that stack serves tens of thousands of sandbox sessions a week, enabling LLMs to take on and complete complex tasks for users. But this is just the beginning. We want someone who can sharpen the architecture that got us here and make the calls that take us further., As an engineer at Adapt, you own systems end-to-end from the customer Slack thread, through architecture and shipping, all the way to the launch blog post (we call it "ship and tell"). Your job isn't done until customers love what you built. As we scale, you'll own whole pillars of the platform. Concretely, you're responsible for: * High-quality execution. This is the central responsibility of any engineer: you design, build, instrument, and confidently scale a system, capability, or feature that just works. Deploys, dashboards, runbooks, and the cost line are part of it, not follow-ups. * Shipping frequently. You ship boulders (like features and capabilities), stones (smaller, quality of life improvements), and pebbles (quick fixes and bugs) regularly and consistently. Infra is no excuse for long dark stretches. * Tying infra work to product outcomes. You can name the customer-visible number a piece of work moves before you start it, and you go check it afterward. * Solving inside the constraints. When you solve a problem by adding resources, you've solved it for your configuration. When you solve it inside the constraints we already have, you've solved it for everyone. Every new code path is a cost. * Customer discovery. You will directly integrate alongside our early customers and the many ahead to understand their pain points of using Adapt and use those to propose and build solutions. * Strong internal and external communication. You will be responsible for writing our customer-centric blog posts and communicating directly with customers when new capabilities launch and how they can be used and understood. * Set a high bar. Lead by example. Expect mentorship, but bring a strong default to figuring things out yourself - and be someone the rest of the team learns from. Especially at the pace of an early-stage startup in a crowded AI market. * Collaboration. You'll work shoulder-to-shoulder with the team and leadership in San Francisco. We bias toward shipping, together. You'll also work directly with leadership: your manager (me, Dustin ), our CTO and co-founder Sean Smith, and the rest of the leadership team. ## Related Videos - [Docker network without Docker](https://www.wearedevelopers.com/videos/1418-docker-network-without-docker) - [Reducing LLM Calls with Vector Search Patterns - Raphael De Lio (Redis)](https://www.wearedevelopers.com/videos/1714-reducing-llm-calls-with-vector-search-patterns-raphael-de-lio-redis) - [Innovating Developer Tools with AI: Insights from GitHub Next](https://www.wearedevelopers.com/videos/1268-innovating-developer-tools-with-ai-insights-from-github-next) - [Shipping Faster with Less: Render on Cloud Hosting, AI Workloads, and the Future of DevOps](https://www.wearedevelopers.com/videos/1894-shipping-faster-with-less-render-on-cloud-hosting-ai-workloads-and-the-future-of-devops) - [Docker exec without Docker](https://www.wearedevelopers.com/videos/1094-docker-exec-without-docker) - [AI Won't Fix Your Engineering Culture](https://www.wearedevelopers.com/videos/100266-ai-won-t-fix-your-engineering-culture) ## Related Articles - [Navigating the AI Shift](https://www.wearedevelopers.com/magazine/629-navigating-the-ai-shift) - [How to Become an AI Engineer](https://www.wearedevelopers.com/magazine/331-how-to-become-an-ai-engineer) - [How We Built a Worry-Free System That Runs for 10+ Years – And What We’d Do Again](https://www.wearedevelopers.com/magazine/751-how-we-built-a-worry-free-system-that-runs-for-10-years-and-what-we-d-do-again) - [What is Software Engineering in the Age of AI?](https://www.wearedevelopers.com/magazine/640-what-is-software-engineering-in-the-age-of-ai) - [Never delegate the understanding](https://www.wearedevelopers.com/magazine/749-never-delegate-the-understanding) - [Dev Digest 120 - Apple and peers](https://www.wearedevelopers.com/magazine/455-dev-digest-120-apple-and-peers)