Hello! We’re a team of performance experts crafting better tools for measuring & improving the performance of computer programs.

We're building Systems Lab, an automated performance testing platform with built-in monitoring & analytics.

Systems Lab makes it easy to understand how your program behaves and how its behavior changes based on service settings, environment configuration, and incoming load.

Have you ever wondered...

💭 What instance size should I use for my app?
💭 How will this code change impact performance in production?
💭 Will updating the OS on our servers cause a performance degradation?

If so, you’ve come to the right place: Systems Lab can help you answer these questions and more.

Our goal is to make performance optimization easy, reproducible, and—dare we say it—fun.

Here’s how it works.

First, get your code running inside of Systems Lab. If your code’s on GitHub, this is as easy as using a GitHub Action.

Next, use one of our templates to ask a performance question. For example, our A/B test template can tell you how changing a system setting will impact your program’s performance, or if there is a performance regression.

Executing a template will conduct a set of performance experiments. Once they’re complete, we’ll help you understand the results in our interactive, visual reporting interface. Of course, full data export is available too.

Unlike most performance tools, Systems Lab presents results in the context of your original question, which makes benchmarks easier to understand and draw conclusions from.

Interested?

Systems Lab currently in the early stages of product development.

Email us if you’re interested in preview access or becoming a design partner to help shape our product direction.

About us

We’re an experienced team of software engineers and performance enthusiasts with a passion for building accessible tools. The team met while working for Twitter, where we saved the company well over $100,000,000 over 5 years.

Yao Yue
Mihir Nanavati
Yuri Vishnevsky
Xi Yang
Brian Martin
Sean Lynch

We have a track record of writing high-performance services and monitoring tools. Here are some of the open-source software projects that we’ve authored and continue to maintain:

Rezolus, a tool for collecting detailed systems performance telemetry and exposing burst patterns through high-resolution measurement.

RPC Perf, a tool for measuring the performance of RPC services.

Pelikan, a cache framework with best-in-class performance and a highly extensible architecture.

If you have questions or comments about what we’re working on, get in touch.