When will sprints take place?
Monday, May 18th, 2026, 8:00am – Tuesday, May 19th, 2025, 9:00pm PST - Sprints are free to attend with a PyCon US registration.
NOTE: Please be familiar with the PyCon US Health & Safety Guidelines. Sprint days are part of the conference and only registered attendees of PyCon US can participate due to the guidelines put in place for the 2026 conference.
Development sprints are a key part of PyCon US, and a chance for the contributors to open-source projects to get together face-to-face for two days of intensive learning, development and camaraderie. Newbies sit with gurus, go out for lunch and dinner together, and have a great time while advancing their project.
Why not join the sprints this year at PyCon US? by Naomi Ceder
What's a sprint?
PyCon US Development Sprints are two days of intensive learning and development on an open source project of your choice, in a team environment. It's a time to come together with colleagues, old and new, to share what you've learned and apply it to an open source project.
In the crucible of a sprint room, teaming with both focus and humor, it's a time to test, fix bugs, add new features, and improve documentation. And it's a time to network, make friends, and build relationships that go beyond the conference.
PyCon US provides the space and infrastructure (network, power, tables & chairs); you bring your skills, humanity, and brainpower (oh! and don't forget your computer).
Who can participate?
You! All experience levels are welcome; sprints are a great opportunity to get connected with, and start contributing to your favorite Python project. Participation in the sprints is free and included in your conference registration. If you are attending sprints, please go to your attendee profile on your dashboard and indicate the number of sprint days you will be attending.
Who can run a sprint?
You! If you've never run a sprint before, the In-Person Event Handbook is an excellent guide.
Instructions for adding a sprint project to this page are below.
What's the schedule?
Sprints run all day from Monday, May 18th through Tuesday, May 19th. That's 8:00am to 9:00pm PST.
Please note: Lunch will not be provided so please plan your lunch breaks accordingly!
Where will the sprints be?
The Sprints will take place at the Long Beach Convention Center. Each Sprinting project will claim its own room or if the room is large enough, it will share the space with other Sprinting Projects.
Which projects are sprinting?
If you are interested in leading a sprint, please add your project using the link below:
Please submit a sprint here and add a brief description of your project that will be listed below. Include links to what you'll be sprinting on. Indicate if the sprint will be newcomer-friendly!
Have questions not covered here?
Reach out to pycon-sprints@python.org
Python Infrastructure
Come work on the Python infrastructure!
BeeWare
Do you want to write an app for your phone using nothing but Python? Have you got some Python code that you'd like to distribute to users as a standalone installer? BeeWare is a collection of libraries and tools for building a cross-platform native graphical user interface using nothing but Python, and then packaging your Python code as a standalone app for distribution on macOS, Windows, Linux, iOS, Android, or as a single-page web app. No matter your level of experience, we can find a way for you to contribute to BeeWare. And every contributor earns a BeeWare Challenge Coin!
hatch
Hatch is a modern extensible environment manager. We support you in your packaging needs with hatchling a batteries included approach to better defaults for packaging your project. We will work on adding new things to hatch, fixing annoying bugs to improve the experience. All skill levels are welcome! We will be holding a sprint on the 18th.
Marcus
Marcus is an open-source orchestration server for AI coding agents. You describe what to build. Marcus breaks the work into tasks on a shared kanban board. Multiple agents pull tasks independently, write the code, and coordinate through the board — never through chat. You walk away; you come back to working software.
Tiny Hub Energy
My project is a DePIN (Decentralized Physical Infrastructure Network) software running on Google Cloud. The IaaS software also features an invisible back end network with plans to batch settlements via Pimlico every 5 minutes. In my opinion, society as a whole will refuse to adapt to anything boasting crypto, we all know it can be a little sketchy. Given all the advancements in AI, a DePIN product can finally have all the automation required and predictive functions needed for major success. Additionally, there is an opportunity here to become the first closed loop cloud software, our cloud compute cost could instantly be replaced with clean energy coming from out customers. Smart home devices and bidirectionally charged EVs are set to create the first real "smart homes" and I envision this orchestrated in real time on the tiny hub network. Microgrid strategies for small communities across the country and a grid upgrade are required to continue to scale this technology. Maybe one day my platform can shoot type 1 energy beams from a space based solar farm straight to a residential home, anything is possible.
PyOpenSci
pyOpenSci supports the Python scientific software ecosystem through its open community-led peer reviews and free Python packaging resources. It also has its own software that requires maintenance. We will work on documentation, infrastructure, security, translations, and other tasks. Ideas for new features are welcome too! This is our fourth year hosting sprints at PyCon US. Read the recap of our 2024 sprint to get an idea what to expect: https://www.pyopensci.org/blog/recap-pyos-pyconus-2024.html#-our-second-pyopensci-sprint- Whether you are an experienced Python developer or just getting started, everyone is welcome!
Arcade
Arcade is an easy to learn library for creating 2D games. It has a user-friendly API that makes it easy for beginners to get started, and expansive lower level utilities that create great possibilities for advanced users. We are fast approaching the release of our new 4.0 version, which brings a lot of under the hood the changes together to bring native support for running in a browser with Pyodide, utilizing WebGL. If you have an interest in game development with Python, check us out! We can always use help even just testing out our pre-release builds of this new version on the web.
GNU Mailman
Mailman is an email list management system complete with a Django based web UI for list management and archiving.
Volute Reasoning Engine (VRE)
What is VRE? VRE gates AI agent tool execution behind a depth-indexed knowledge graph. Before any tool runs, VRE checks whether the agent can structurally justify its understanding of the operation across four depth levels: existence, identity, capabilities, and constraints. If understanding is insufficient, execution is blocked and the agent is told exactly what it doesn't know. The enforcement lives in a decorator, not a prompt. It can't be bypassed by reasoning, lost during context compaction, or ignored by a capable model. Agents cannot act on knowledge they cannot justify. Why this matters AI agents routinely present inference as fact, execute operations they don't understand, and confuse authorization with comprehension. Current safety approaches (system prompts, guardrail classifiers, LLM-as-judge) are linguistic and degrade over conversation length. VRE's enforcement is structural and operates outside the context window entirely. Sprint Goals All experience levels welcome.- Domain graphs: VRE ships with a filesystem graph. We need git, Docker, Kubernetes, PostgreSQL, and more. If you know a domain, you can author its epistemic model. Domain expertise required, Python optional.
- Tool plugins: Each plugin bundles a tool implementation, its domain graph, and a concept extractor. Good fit for anyone familiar with LangChain or agent tooling.
- Harness TUI: Building Voli, a terminal interface for running local agents with VRE enforcement, using
textual. - Docs and examples: Tutorials, integration guides, and standalone examples.
ScanAPI
ScanAPI 🪼 ScanAPI is an open source API integration testing framework built in Python and maintained by Cumbuca Dev, a Brazilian FLOSS community focused on contributor onboarding, open source sustainability and collaborative development. The project provides:- Automated Integration Testing
- Automated Live Documentation
- Real-time API diagnostics
- Flexible YAML/JSON-based configuration
- Contributor-friendly architecture for extensibility and experimentation
Memray
Memray is a Python memory profiler - see our docs. We'll be working on adding some new features, including Python 3.15 support. Many of the open issues require some C or C++ knowledge to complete, though there are a few that only require Python or JavaScript. Feel free to join us whether you'd like to contribute a new feature, or just learn more about Memray or get help trying it out!