Software Engineer - Platform

We built our own time-series database. Existing engines could not handle our workload. Ours stores petabytes at nanosecond precision. It ingests millions of points per second. It answers queries over trillions of points in milliseconds. The code is open source.

That database powers our platform. Our customers run critical infrastructure across North America. One customer can produce a million measurement points per second. For us, that is normal.

We are a small team. We are venture-backed and profitable.

What you will work on

You work across the whole platform, not one slice of it.

• Storage engine. On-disk layout, compression, tree structures, and correctness.
• Ingest. Thousands of sensors. Clock skew, late backfill, duplicate streams, and dropouts.
• Query and APIs. Give users the power of the engine. Hide the complexity.
• Production. You run what you build. Kubernetes on AWS, GovCloud, and Azure. Our customers have no maintenance window.

What we look for

We want depth. Somewhere in computing, you know one corner far better than any job required. Maybe you wrote a toy database, a compiler, an emulator, or a scheduler. You did it because you wanted to know how it worked.

• Strong Go. Or strong systems work in C, C++, Rust, or Zig, plus real interest in Go.
• Real knowledge of database and storage internals. On-disk layout, compaction, indexing, caching, and write amplification.
• Distributed systems fundamentals. Concurrency, consistency, failure modes, and networking.
• Evidence that you can operate what you build. Profiling, tracing, and incident response.

Nice to have: Rust. Prior work on a database, query engine, or storage layer. Performance engineering. High-throughput streaming. Kubernetes internals. Sensor-heavy domains.

Not required: A CS degree. A set number of years. Domain experience.

How we work with AI

AI is a multiplier on judgment, not a replacement for engineers. We ask everyone to use it every day for generation, review, debugging, and tests. We ask you to experiment and to share what works. We are also honest about the limits. Models are confidently wrong, and that is expensive at petabyte scale. That is why we hire people who can tell when the answer is wrong.

Perks

• Fully remote
• Equity
• Annual bonus
• Full benefits
• Small team, real ownership, deep technical work

How to apply

Send your CV to time.series.database@golang.cafe.