OpenReproLab teaches first-year PhD students the industry-grade computing tools and methods to produce FAIR and reproducible science. More generally, this programme fosters a culture of scientific openness and transparency, where all relevant data and software produced during the scientific activity are freely accessible and reusable, well-documented, and citable, and where all published results are fully reproducible.
We acknowledge that computing skills are best learned when working on real projects. OpenReproLab will be structured around lab sessions ("travaux pratiques") where participants will solve practical problems related to producing FAIR and reproducible science.
During lab sessions, which are similar in design to data/coding challenges, each participant will work on practical problems in the presence of one or more instructors to answer their questions and guide their learning. Lab sessions may involve, for example, writing computer programs, configuring computing environments, and writing documentation or metadata. Students will submit their work via git commits on a git-hosting online platform. Real-time automated feedback will be given to students via the platform, so that they can validate their work, learn from their mistakes, and keep trying until they get it right. All necessary material will be provided for each lab session but, to the extent it is possible, participants will be given the option to use their own PhD research material instead (eg. their own data sets).
Volume: Seven 2-hour lab sessions (summary lecture + practice), cf. programme below
Each lab session will start with a summary lecture on the day’s relevant topics.
Volume: The first 30-45 minutes of each lab session.
Prior to each lab session, participants will be asked to acquire knowledge autonomously on the topic at hand via readings and/or online tutorial videos. For each topic, participants will be given a one-page summary of the key points and take-home messages, along with the list of readings/videos to prepare for the corresponding lab session.
Volume: About two hours of autonomous work prior to each lab session.
In order to encourage and facilitate the autonomous work required of participants prior to each lab, we will organize optional study sessions where students can gather up to do the work together. These sessions will be unsupervised (no instructor present) so as to make it a safe space for students to interact among peers.
Volume: Two hours of study session, one week before each lab session.
We will hold regular office hours, where students can come without appointment to talk one-on-one with instructors about any topic relevant to OpenReproLab, including asking for clarification on past material, asking for help on unfinished lab sessions, discussing more advanced topics, and how to apply OpenReproLab principles to their own research.
Volume: Two hours (one instructor) per month.
A few plenary meetings will be necessary for organizational purposes, such as a first "welcome" meeting, to greet everyone and present the programme.
Volume: Three 1-hour meeting (welcome meeting + intermediate meetings + final meeting).
Social events with optional attendance will be organised to foster cohort cohesion and facilitate long-lasting professional connections among participants. These will be held on campus, in the mid- to late-afternoon. IGE will provide snacks and soft drinks.
Volume: Two social events
-
First steps in numerical science: Git basics, writing a Python script, writing a Jupyter Notebook
-
Data management: metadata and documentation, traceability, CF conventions, DOI, versioning, data warehouses.
-
Code management, part 1: licensing, modularity, documentation.
-
Code management, part 2: managing dependencies and computing environments.
-
Team-oriented development: leveraging Git and online Git-hosting platforms to develop software as a team, git branching, pull requests and code reviews, documentation, governance.
-
Reproducibility challenge: try to reproduce results from a real-life scientific paper, analyse what limits reproducibility, propose solutions.
-
Open science and artificial intelligence: using large language models to produce scientific code: use cases, recommendations, risks and pitfalls, accountability.
OpenReproLab will accept up to 25 participants. Are eligible, from highest to lowest priority:
-
IGE PhD students who started in calendar year 2026
-
IGE PhD students who started in calendar year 2025
-
IGE IE/IR BAP E employed as CDD
For PhD students, this programme is worth 13 hours among the 120 hours of mandatory training. We will deliver an attestation of training completion to students who have attended 6 out of the 7 lab sessions in person and who have completed all the work for the 7 lab sessions. This work can be finished after the lab sessions have ended, but no later than May 20th.
We intend to make OpenReproLab accessible to everyone. If you have a disability that makes this programme less accessible to you, please let us know as soon as possible, so we can implement relevant accommodations. We respect your privacy and we will not share any disability-related information outside of the OpenReproLab organisers.
OpenReproLab is organised by IGE’s plateforme d’innovation numérique en géosciences (PING).
Unless otherwise noted, all events will take place at Maison Climat Planète (MCP), on the UGA campus.
Plenary meetings and lab sessions will take place in room 117/118 according to the following schedule:
| Date | Session |
|---|---|
| Thursday October 15th 2026, 10am-11am | Plenary meeting (presentation of ORL) |
| Thursday November 19th 2026, 10am-12pm | Lab 1 |
| Thursday December 17th 2026, 10am-12pm | Lab 2 (in IMAG building, room 2) |
| Thursday January 14th 2027, 10am-12pm | Lab 3 |
| Thursday January 28th 2027, 10am-11am | Plenary meeting (discussing ORL so far) |
| Thursday February 11th 2027, 10am-12pm | Lab 4 |
| Thursday March 11th 2027, 10am-12pm | Lab 5 |
| Thursday April 8th 2027, 10am-12pm | Lab 6 |
| Thursday May 13th 2027, 10am-12pm | Lab 7 |
| Thursday May 27th 2027, 10am-11am | Plenary meeting (wrap up of ORL) |
Optional instructor-free study sessions will take place on the Tuesday immediately prior to each lab session, same room, same time (117/118, 10am-12pm).
Office hours will take place in room 200 according to the following schedule:
- Monday December 7th 2026, 2pm-4pm
- Monday January 4th 2027, 2pm-4pm
- Monday February 1st 2027, 2pm-4pm
- Monday March 1st 2027, 2pm-4pm
- Monday March 29th 2027, 2pm-4pm
- Monday May 3rd 2027, 2pm-4pm
Social events will take place starting at 5pm:
- In the MCP cafeteria on Thursday January 14th 2027
- On the MCP rooftop on Thursday May 27th 2027
Copyright (C) 2026 Institut des Géosciences de l'Environnement, Grenoble.
The computer code of this repository is released under the terms of the GNU General Public License v3.0.
The rest of the content of this repository (including this document) is released under the terms of the Creative Commons Attribution Share Alike 4.0 International license.