# SciPy2025 Sprints

Please add your project and who is leading/the contact person for the sprint. And maybe add when you will leave today or tomorrow. (e.g. sprint goes on until Sunday noon, etc.).

Please keep the same order of sign-up and pitching :slightly_smiling_face:

Link to this docs: https://hackmd.io/@bsipocz/scipy2025_sprints

Use "University of Washington" wifi.

[toc]

## Rooms

All rooms have whiteboards, please put your project name on a board placed at the entrance of the room. You are welcome to move the teams around, but try to keep the list below up-to-date to help people find you.

### Hall

* napari
* xarray
* zarr
* matplotlib
* DragonHPC
*

### 105

* pyOpenSci
*
*

### 106

*

### 108

* NumPy
* scikit-build-core
* CuPy / CUDA Python
* Astropy
*

## Projects

### NumPy (Room 108)
* Contacts:
* Nathan Goldbaum (@ngoldbaum)

* Github link
* https://github.com/numpy/numpy
* https://github.com/numpy/numpy/issues?q=is%3Aissue%20state%3Aopen%20label%3A%22sprintable%22
* https://github.com/numpy/numpy/issues?q=is%3Aissue%20state%3Aopen%20label%3A%22sprintable%20-%20C%22

* What is it about
Do you want to learn how to debug and understand the C internals of NumPy so it is less of a black box?
Is there a particularly annoying NumPy you would like to see fixed?
Just want to get involved?

We will be sprinting on NumPy. I will personally be focusing on topics related to supporting free-threaded Python, but am happy to help out with any topics anyone wants to work on.

We have `sprintable` and `sprintable - C` labels for issues. The former require no C knowledge while the latter can be tackled if you have experience with C but not NumPy in particular.

### Scikit-build-core (Room 108)

* Contacts:
* Henry Schreiner @henryiii

* GitHub link
* https://github.com/scikit-build/scikit-build-core
* What is it about
Do you have a compiled package you'd like to have a nice build system for? Support new platforms like WebAssembly or iOS? We can work on helping you transition from something like a setup.py to scikit-build-core. Last time we held this sprint at SciPy someone moved from 800+ line setup.py to <20 lines.

If you are already using scikit-build-core, this is a chance for us to work on adding or fixing anything you thing could be better.

(We can also look at issues in @henryiii's other projects, like cibuildwheel, Scientific Python Development Guide, nox, etc. if you want!)

### LlamaBot (Hall)

* Contacts:
* Eric Ma
* https://github.com/ericmjl/llamabot
* What it's about:
* Check out the #scipy-sprints tag on https://github.com/ericmjl/llamabot/issues
* Contributions to documentation
* If you're curious to build something with LlamaBot, please feel free to swing by and ideate with each other.

### napari

We're here all weekend!

* Contacts
* Peter Sobolewski (@psobolewskiPhD)
* Tim Monko (@TimMonko)
* Draga Doncila Pop (@DragaDoncila)

* GitHub Link
* github.com/napari/napari

* Getting Started
* Check out our quick intro doc or come say hi!
* https://hackmd.io/@napari/scipy-2025-napari

* What's going on?
* napari :handshake: xarray
* We are here to support you with making your first contribution to napari!
* if you have image/image-like data, we can help you load it and add it as an example to our gallery
* you can improve our documentation
* you can contribute to the core napari viewer
* We're also happy to support with your own analysis, if you have data you'd like to play with or a particular workflow you'd like to build
* We can also work with you to build a plugin!
* Reach out if any of this sounds interesting. We're very friendly, we promise!

### CUDA Python

* Contacts:
* Katrina Riehl ([e-mail])
* Andy Terrel
* What's going on?
* We are here to answer questions about CUDA Python, RAPIDS, and DevTools.
* We can pair program with you on GPU-accelerated backends and workflows.
* Working through Educational Materials in our GH repo: https://github.com/NVIDIA/accelerated-computing-hub
* CuPy Sprints (Leo Fang)

### Shiny: Dashboarding in Python (ROOM 105)

* Contacts:
* Daniel Chen ([e-mail])

- Scipy tutorial: <https://github.com/chendaniely/scipy-2025-shiny>
- Shiny github: <https://github.com/posit-dev/py-shiny>

If you attended/missed the tutorial and you want to get a dashboard up and running (maybe even deployed???). I'll be here to help you get started and learn more about Shiny.

### Chatlas + Querychat: talk to your data with LLMs in Shiny (ROOM 105)

* Contacts:
* Daniel Chen ([e-mail])

If you missed or want to learn more about Joe's talk about putting guardrails around LLMs for data science and research and want to get started with your own project. I'm here to help you get an example data chat app working for you

- schedule: <https://cfp.scipy.org/scipy2025/talk/WM9UFJ/>
- Talk repo: <https://github.com/jcheng5/SciPy-2025>
- chatlas repo: <https://github.com/posit-dev/chatlas>
- querychat repo: <https://github.com/posit-dev/querychat>
- docs: <https://posit-dev.github.io/querychat/>

### Quarto: Reproducible documents (ROOM 105)

* Contacts:
* Daniel Chen ([e-mail])

If you missed or want to learn more about Charlotte Wickham's [From One Notebook to Many Reports: Automating with Quarto](https://cfp.scipy.org/scipy2025/talk/LNWCSE/) talk.
I'll be here to help you with your quarto questions and help building technical artifacts using Quarto + Python

- schecule: <https://cfp.scipy.org/scipy2025/talk/LNWCSE/>
- repo: <https://github.com/quarto-dev/quarto-cli>
- docs: <https://quarto.org/>

### pyOpenSci ROOM 105 :sparkles:

* Contact: Leah Wasser ([e-mail])
* You can also connect with us and participate remotely via our [slack channel](https://join.slack.com/t/pyopensci/shared_invite/zt-38rexn7a8-fxk6Q6pEfyJX9iMHBN73DA)
* I am only here today :)
* [Here is our project board with help-wanted / sprintable tasks](https://github.com/orgs/pyOpenSci/projects/3)
* Things you can help with
* Work though our packaging tutorials and help us fix bugs / issues
* Translate our packaging guide (see above!) to Spanish/ japanese OR add a new language :sparkles:
* Go through our [packaging tutorials](https://www.pyopensci.org/python-package-guide/tutorials/intro.html), test them out, open issues and find problems / points of confusion / etc and lets fix them together
* Go through our new contribute to [open source lessons](https://www.pyopensci.org/lessons/contribute-open-source/your-first-contribution.html#pyos-first-contribution), learn about contribution workflows and submit issues and pr's to fix problems in those lessons :)
* Help us fix infrastructure issues (python / ci across our organization - [all listed on our project board](https://github.com/orgs/pyOpenSci/projects/3/views/1))
* [Tetsuo Koyama](https://github.com/tkoyama010) <- :) :heart:
* If you'd like to translate the document into your language, I can support you online until my flight at 1 p.m. today and also tomorrow.
* If you're interested in the [PyVista project](https://github.com/pyvista/pyvista), I'll be sprinting online during the same schedule as the pyOpenSci project.

Contriubte to open source lessons
https://www.pyopensci.org/lessons/contribute-open-source/your-first-contribution.html#pyos-first-contribution

### Matplotlib (Rm 105)

* Contact
* Kyle Sunden (@ksunden)
* Github link
* https://github.com/matplotlib/matplotlib
* https://github.com/matplotlib/matplotlib/issues?q=state%3Aopen%20label%3A%22Good%20first%20issue%22

### DragonHPC

* Contact
* Davin Potts (@applio)
* Colin Wahl
* Github link
* https://github.com/DragonHPC/dragon/
* What is this?
* Dragon lets your `multiprocessing` code execute across multiple nodes without any changes to code beyond importing Dragon.
* Poster
* Dragon also provides an in-memory distributed dict (`DDict`) that can be used to back Zarr, dramatically speeding access as demonstrated with OME-Zarr microscopy datasets and CZBio's use cases around napari
* bit.ly/dragonhpc (link to notebook with content from poster)
* What could we do together?
* Get anyone that wants help up and running with Dragon
* New to `multiprocessing`? `concurrent.futures`? Parallel things that use those Python Standard Library modules? Happy to help there too.
* Help improve some of Dragon's docs, adding examples, etc.
* Help test the Dragon executor for Cubed

### Astropy & friends

* Contact
* Matt Craig (@mwcraig)
* What this about

There are several open issues in astropy-related projects. I'm happy to help you find one to fix that is the right scale for. I'm also looking for help on a couple of specific things that not simply issues.

* Places to look for issues
* List of [affiliated packages](https://www.astropy.org/affiliated/index.html#coordinated-package-list) -- open issues here are more approachable than issues in astropy itself.
* astropy core: https://github.com/astropy/astropy
* Some more niche projects in need of some love, including better documentation:
* AIDA (Astro Image Display API): https://github.com/astropy/astro-image-display-api
* astrowidgets (backends for AIDA): https://github.com/astropy/astrowidgets
* stellarphot: https://github.com/feder-observatory/stellarphot
* Help wanted
* I need to test a package on GPUs -- help getting that set up on Brev would be awesome.
* If you maintain a plotting package I'd be interested in talking to you about implementing a backend for AIDA.

### Numba v2 + Egraph (egg/egglog) (ROOM 105)

(Btw, I brought home-made cookies!)

Contact:

- Siu Kwan Lam (GH: sklam)

What is this about?

- Numba v2 uses a new technique called Egraph
- The work is done as a book.
- Come learn about compilers
- Come learn about how your project can use EGraph

Help wanted:

- Help reviewing the book.
- The book is new. We don't know if it is easy to read.
- A chance to learn about compilers

Links:
- The "Baby Dragon Book": https://numba.pydata.org/numba-prototypes/sealir_tutorials/index.html
- Github: https://github.com/numba/numba-prototypes/tree/main/sealir-tutorials
- Scipy Proceeding on Numba's SuperOptimization technique and EGraph extraction: https://github.com/scipy-conference/scipy_proceedings/pull/1094

### Xarray/Zarr/Icechunk/Cubed

Contact:
- Deepak Cherian (@dcherian, [e-mail])
- Ian Hunt-Isaak (@ianhi)
- Justus Magin (@keewis)

What is this about?
- Xarray for biological subfields
- integration with napari et al?
- examples for working with mciroscopy/neuro/genomics etc. data in xarray?
- Xarray for astro?
- represent complex coordinate systems in Xarray
- Experiment with Zarr & Icechunk?
- Integration of marray and xarray
- are you looking to experiment with xarray?

Links:
- https://xarray.dev/blog/xarray-biology
- https://xarray-indexes.readthedocs.io/

### Extralit

Fast and accurate systemic literature data extraction with LLM assistance in a data annotation framework

GH Repo: https://github.com/extralit/extralit

Build a papers library import flow to integrate with Zotero and Mendeley collectons.

Tech stack:
- pymupdf
- Vue.js
- Pandera
- LlamaIndex

Contact:

- Jonny Tran ([e-mail])

### pyHPC

- rebuilding the Python Community at Supercomputing
- write proposal for a BOF
- setup website and email

github: https://github.com/pyhpc

Contact:
- Andy Terrel

### Discover Cookbook

- rebooted project in dec
- trying to get a third edition by end of the year

github: https://github.com/numfocus/DISCOVER-Cookbook/

Contact:
- Andy Terrel

### Pixi x marimo

- getting

### Oxbow

* Contact
* Nezar Abdennur (@nvictus)
* Conrad Bzura (@conradbzura)
* GitHub link
* https://github.com/abdenlab/oxbow
* Docs
* https://oxbow.readthedocs.io
* What it's about?
* Oxbow translates 11 specialized genomic file formats into Apache Arrow for use in distributed and streaming analytics libraries like Polars, Pandas, DuckDB, and Dask.
* We'd love to release wasm-emscripten oxbow wheels for use in pyodide environments like Jupyter-lite and Marimo! Some issues because Oxbow is a Rust+Python project.