Senior Platform Engineer (CI/CD & Developer Experience)

DuckLabs is the company behind the popular DuckDB database system. We build and distribute dozens of DuckDB clients for all major operating systems and hardware platforms. What’s more, DuckDB has a large extension ecosystem, which consists of the core extensions developed by the DuckLabs team, and the community extensions submitted by third parties.

All of this means that we have a complex Continuous Integration pipeline with hundreds of different jobs, along with a steady stream of contributions to build, review, test and integrate into our codebase. Therefore, we are looking for a second Senior Platform Engineer to join the team and work on the tooling behind DuckDB. You’ll be in a critical position in the organization enabling all of Engineering and helping shape what our CI pipelines look like from here.

About us

DuckDB is a modern analytical database that can run in-process, i.e., inside the client application. DuckDB is highly portable: it runs on your laptop, on a server or in the browser, and it gets downloaded more than three million times a day. It started in the Database Architectures group at CWI in Amsterdam. DuckLabs is the company that builds it, along with DuckLake (our lakehouse format) and Quack (a remote protocol that turns DuckDB into a client-server database).

You might have seen the news: in August, we joined Amazon Web Services. Since then, we’ve operated as a subsidiary, but we’re staying in Amsterdam, the founding team is still running things, and DuckDB stays open source with the MIT license under the DuckDB Foundation.

The team

We’re around 30 people, mostly engineers. It’s a tight-knit group and we still do one stand-up with the whole engineering team. People here work across the stack instead of sticking to one corner of it, and that goes for this role too. Platform Engineering is currently a team of one, Sander, and you’d make it two. You’ll split the work by what each of you is strongest at, with enough overlap to cover for each other when needed.

What you’ll be doing

Mostly, making sure our engineers can keep shipping while the amount of change keeps growing. In practice, that means:

• Taking a hard look at our CI/CD. We help engineers to create software that is correct, fast and reliable on Linux, Windows, macOS and WebAssembly. We want to balance developer velocity with finding errors in code changes. We decide what to test, how, and at which stage of development. This feedback loop needs to be accurate, fast and reliable.
• Bringing the cloud into CI. We want test environments and performance tests that run against real backends like S3 and Iceberg catalogs, with numbers we can trust. We help engineers measure the impact of their changes on the network, disk, CPU and memory performance for single processes as well as complex setups involving many services.
• Automating how we work. We want issue trackers, PRs and CI linked up so the toil disappears and the routine work runs itself.
• Looking after the toolchain and releases: compilers, build systems, build speed and the release process. When something won’t compile on Windows or a linker throws an error on Linux, you’ll often be the one who figures out why.
• Growing our extension ecosystem. Extensions are a big part of what makes DuckDB useful, and there are more every month. You’ll help shape how that scales: clear versioning, explicit dependencies between extensions, streamlining community extension contributions, etc.

Our stack

Our CI runs on GitHub Actions, Namespace and Cloudflare, with CMake, vcpkg and Python. DuckDB itself is mostly written in C++, with client libraries in Python, Go, Rust, Java, JavaScript and more.

What we’re looking for

• You work autonomously and naturally take ownership. Things move fast nowadays, and to keep up we need all our engineers to be critical, independent thinkers. You’ve had the technical lead on large components or processes, not just carried out someone else’s plan.
• A good few years as a platform engineer, SRE or build/release engineer, on systems lots of other engineers depended on.
• CI/CD at real scale, ideally with GitHub Actions. You know what it’s like when the merge queue is stuck and nobody can merge anything.
• Cloud infrastructure automation, preferably on AWS. Setting up test environments as code shouldn’t be new to you.
• Comfort with compiled code. You know how compilers, linkers and CMake fit together, and you don’t mind debugging on Windows and macOS. You don’t need to be a C++ expert; we have plenty of those.

Nobody ticks every box. If you’re excellent in one of these areas, say cloud, builds or developer tooling, we’d like to hear from you.

Nice to have

• Having led a platform, SRE or build team
• You’ve built your own developer tooling: local dev setups, linters, faster feedback loops
• Release management
• Security or DevSecOps
• Setting up agentic coding harnesses
• Cross-compilation, package managers, or more than one cloud
• Maintaining an open-source project or making significant contributions to one

Good to know before you apply

This is real engineering work: the people you’re building for are our own engineers and any other engineer working with our Duck Stack. If you do it well, they ship faster. It also moves quickly. You usually only see the next problem once you’ve solved the one in front of you, so don’t expect a neat roadmap. We read every CV ourselves, and the bar is high. Also, we’re in the office at least three days a week, at Amsterdam Science Park.

What we offer

• An Amazon/AWS-level salary, based on your experience
• High-impact work that ends up in front of millions of developers
• A small team that cares about deep work, not long hours
• 25 vacation days
• Pension plan

Interested?

Ready to build the (build) engine behind the database? Awesome, let’s have a chat!