Threadr builds libraries that make concurrency easier, pipelines cleaner, and engineering a little more fun. Everything here is open-source, actively maintained, and shaped by real-world needs.
These are the tools we reach for when we want systems to feel fast, safe, and deeply satisfying to work with.
Libraries
Open-source projects born from client work, research spikes, and stubborn curiosity.
A high-performance Go library for building async pipelines, fan-out/fan-in workflows, and worker pools without the footguns.
Why it exists
To make concurrency safe, expressive, and actually pleasant.
Features include
The Python counterpart to gliter — an ergonomic async pipeline library for I/O-bound workloads, multiprocessing, and complex dataflows.
Why it exists
To give Python a concurrency story developers actually enjoy.
Features include
A community-powered certification for “blazingly fast” OSS — part joke, part movement, all good vibes.
Why it exists
To celebrate performance culture, humor, and the joy of building in public.
What it offers
A lightweight dataframe library with a small, pleasant API for working with rows and columns without drowning in abstraction.
Why it exists
To give Rust developers a simple, fast way to manipulate row/column structures without ceremony or bloat.
Features include
Principles
Defaults that show up across every repo, README, and PR.
APIs should guide you toward correct parallelism, not punish mistakes or add fear.
If a pipeline is hard to extend, it won’t survive real workloads. Ergonomics first.
No magic. Just patterns you can trust — even at 3 AM.
Tools land better when they bring people together. Shipping is the start of the conversation.
Ready to build?
Let’s architect something fast, reliable, and resilient enough to survive real-world load without weekend pages.