QuantStack
@quantstack.bsky.social
A team of open-source maintainers and engineers providing solutions and support for open-source projects used by millions worldwide. #Jupyter, #condaforge, #Mamba, #ApacheArrow, #Xtensor, and more.
JupyterLite 0.8 is released! 🎉 This release updates to the latest JupyterLab 4.6 and Notebook 7.6, and brings new improvements across related projects. ➡️ Try it online: jupyter.org/try-jupyter ➡️ Blog post: blog.jupyter.org/jupyterlite-...
JupyterLite 0.8 is released! 🎉
JupyterLite is a Jupyter distribution that runs entirely in the web browser without any server components. Deployment and hosting are easy…
blog.jupyter.org
Congratulations to the newest batch of Jupyter Distinguished Contributors, James Colliander, Florence Haudin, Gregory Mooney, and Kirstie Whitaker. The Jupyter community thanks you for your effort to advance Project Jupyter, and for leading by example! blog.jupyter.org/congratulati...
Congratulations, Distinguished Contributors!
We are proud to announce the recipients of the Jupyter Distinguished Contributor (JDC) award for the 2025 cohort of contributors.
blog.jupyter.org
Journées Open Source de @insee.fr au lieu de la transformation - @sylvaincorlay.bsky.social de QuantStack fait la keynote d'ouverture, reliant son expérience d'entrepreneur dans l'open source et sa participation (et celle de son entreprise) au projet @jupyter.org
New SIMD improvements make Parquet column reads up to 80% faster on baseline ARM64 and SVE. Shipping with @arrow.apache.org 25.0. Work done @quantstack.bsky.social with support from a generous sponsor. medium.com/@AntoineProu...
Compute! Paris 2026 Sprints on Nov 27! We’re excited to announce a full day of coding, collaboration, and open-source contributions in scientific computing. Hosted by SCAI, our community partner. scai.sorbonne-universite.fr Learn more: compute.events/paris2026/sprints.html
⚠️There are only a few days left to submit your proposal to Compute! Paris 2026. 🗓️ Cfp closes on Sunday (June 7th). 🔗 compute.events/paris2026/
Compute! Paris 2026
Understanding your tools and algorithms is what gives you agency in a digital world. Paris, November 25–26, 2026.
compute.events
Wolf Vollprecht, founder of Prefix.dev, creator of mamba & pixi, and conda-forge core maintainer, is keynoting Compute! Paris 2026! Hear his vision for the future of open-source tooling, performance, and data science infrastructure.
Mackenzie Mathis (EPFL neuroscientist, DeepLabCut & CEBRA co-creator and open science pioneer) is keynoting at Compute! Paris 2026! Her tools (used by 1000s of labs) decode behavior & neural activity. She explores adaptive intelligence, how brains and machines learn in a changing world.
We added top-level await support to xeus-python 🚀 You can now use toplevel async code directly in notebooks without extra boilerplate — and without blocking the kernel. Available with xeus-python 0.19.0 Here is a short blog post about how it works: notebook.link/@DerThorsten...
Notebook.link - Run & Share Jupyter Notebooks Instantly
A fully interactive, scalable, and language-agnostic computing environment that operates entirely in your browser. Supports Python, R, C++, and GNU Octave. Share a link — anyone can run it instantly.
notebook.link
📣 We are delighted to release mamba 2.6.0. 📣 This minor version includes the support for sharded repodata, significantly improving the UX of the project. 🔗 Read more about improvements on this blog post: medium.com/@julien.jerp... 🔗 For a full overview for the change: github.com/mamba-org/ma...
Sharded Repodata Support in Mamba
Introduction: conda packages metadata indexing
medium.com
Biologists, city planners, physicists, and engineers... What do they have in common? At Compute! Paris 2026 (Nov 25-26), they’re solving the same data challenges from different angles. Join the conversation! compute.events/paris2026/ #ComputeParis
Compute! Paris 2026
Understanding your tools and algorithms is what gives you agency in a digital world. Paris, November 25–26, 2026.
compute.events
Working on a new feature for the next JupyterLab 4.6 release that makes it easier to move panels and tabs across the interface and customize the layout to your liking. This is going to be useful for working with your favorite coding agent CLI right from within JupyterLab! 🤖
Compute! is a new conference series built on the idea that data professions need transparency, not black boxes. It connects open-source maintainers with users, fostering collaboration across disciplines. The first event will happen in Paris in November, and the CfP is open! compute.events/paris2026/
Compute! Paris 2026
Understanding your tools and algorithms is what gives you agency in a digital world. Paris, November 25–26, 2026.
compute.events
Explore the expansion of geospatial workflows in Jupyter. Gregory Mooney, QuantStack highlights how JupyterGIS is enabling collaborative mapping and spatial analysis within interactive notebooks. #Jupyter #GIS #OpenSource blog.jupyter.org/expanding-ge...
Expanding Geospatial Workflows in JupyterGIS: STAC Browsing and Story Maps
Since its initial announcement less than two years ago, JupyterGIS has been a growing effort to bring interactive geospatial workflows into…
blog.jupyter.org
At QuantStack, we are deeply committed to advancing the @arrow.apache.org ecosystem. Today, we introduce Sparrow-IPC, a modern C++20 implementation of Arrow IPC by Hind Montassif, Alexis Placet @alexis-placet.bsky.social, and Johan Mabille. More details in our post: medium.com/@QuantStack/...
Introducing Sparrow-IPC: A modern C++ implementation of Arrow IPC
If you work in the world of data engineering, you are probably familiar with Apache Arrow — the universal columnar format and…
medium.com
The notebook.link team is building "serverless scientific computing" on top of the modern Conda ecosystem & the prefix.dev package server. Read the guest blog on how they shipped interactive WASM for Python, R, C++, Fortran and more: prefix.dev/blog/server...
How notebook.link Scales Without Backends
Notebook.link combines WebAssembly and the conda ecosystem to deliver scalable, serverless computing environments that run entirely in the browser. No backend. No scaling headaches. Just instant, collaborative scientific computing for everyone.
prefix.dev
Dive into our rewrite of a low-level algorithm in @arrow.apache.org using C++ metaprogramming that improves the performance of #Parquet reading by up to 50-60%. Impressive work done mostly by @antoineprv.bsky.social 1/ Highly understandable overview at medium.com/@AntoineProu...
Faster Reads for Apache Parquet: Improving Integer Unpacking
By rewriting the SIMD optimizations of a critical low-level algorithm in Arrow C++, we have been able to deliver substantial speedups when…
medium.com
Recreating Ed Hawkins' climate spiral visualization in a Jupyter Notebook on notebook.link 🎨 The notebook uses NASA GISTEMP v4 monthly temperature anomalies from 1980 to 2025, and the p5.js JupyterLite kernel for ad-hoc visualizations 💡 ➡️ Try it in your browser: notebook.link/@jtp/paintin...
Backstage interview: @ylechelle.bsky.social Ian explained how :probabl turned scikit-learn (a widely used open-source tool) into a sustainable business while keeping it open. His message: Open source is Europe’s best shot at tech independence, from open data to open hardware. youtu.be/VMOpvJei4hk
Backstage interview - Yann Lechelle
www.pydata.org PyData is an educational program of NumFOCUS, a 501(c)3 non-profit organization in the United States. PyData provides a forum for the international community of users and developers…
youtu.be
And this was all done by @antoineprv.bsky.social , whose BlueSky account isn't very discoverable. 😊
At @quantstack.bsky.social we designed novel bit-unpacking SIMD optimizations for @arrow.apache.org and #ApacheParquet, and implemented them entirely using C++ metaprogramming instead of Python-based code generation. We'll publish a deep dive blog post soon. github.com/apache/arrow...
GH-48277: [C++][Parquet] unpack with shuffle algorithm by AntoinePrv · Pull Request #47994 · apache/arrow
Rationale for this change The current bit-unpacking algorithm (which is implemented as a C++ code generator script in Python) does not fully leverage SIMD operations: all loads and some bitshifts u...
github.com
@geojupyter.bsky.social 's JupyterGIS just got two new features: built-in STAC browsing and Story Maps. Discover #geospatial data and tell map-based stories all from @jupyter.org! Try it live in your browser: notebook.link/@quantstack/.... Read more on the Project Jupyter Blog: shorturl.at/UXE8s
Expanding Geospatial Workflows in JupyterGIS: STAC Browsing and Story Maps
Since its initial announcement less than two years ago, JupyterGIS has been a growing effort to bring interactive geospatial workflows into…
shorturl.at
🧪 DataLab-Kernel | Advanced plotting showcase Running in the browser via notebook.link from @quantstack.bsky.social The plotter now implements all core DataLab visualization features Standalone demo — no DataLab required DataLab-Kernel: github.com/DataLab-Plat... DataLab: datalab-platform.com
🎉 JupyterLite is now part of Project Jupyter! Run Jupyter notebooks directly in your browser: no server, no setup. Many thanks to the JupyterLite contributors and supporters like QuantStack, Bloomberg and the Gates Foundation. 🔗 blog.jupyter.org/jupyterlite-...
JupyterLite Officially Joins Project Jupyter!
We are thrilled to announce that JupyterLite is now an official part of Project Jupyter. This milestone marks a significant step forward…
blog.jupyter.org
🧪 Sigima, running in the browser thanks to notebook.link from @quantstack.bsky.social The scientific signal & image processing engine behind DataLab (datalab-platform.com), now executable directly in a @jupyter.org notebook (no local install required). Lightweight, reproducible, easy to share.
🧪 Awesome announcement from the @quantstack.bsky.social team! Introducing notebook.link, share your computational notebook and have them run directly in the browser with no setup, powered by Jupyterlite and the xeus stack. #datascience #scientificcomputing
Introducing notebook.link: The Future of Notebook Sharing
Reimagining how you share, collaborate, and run Jupyter notebooks — all in your browser.
medium.com
🧪 Live experiment using notebook.link from @quantstack.bsky.social A notebook with DataLab-Kernel (github.com/DataLab-Plat...), a Xeus-Python JupyterLite-compatible kernel, connects automatically to a running local DataLab app (datalab-platform.com). No config. Auto-discovery. It simply works.
