• Attending
• News
• Joint WLCG/HSF workshop 2026 in Bologna
  • AG meeting at the workshop
• Activity Area activities
• WLCG Management Board community software liaison
• Advisory Group news and follow-ups
• “Responsible AI-assisted coding and common training aspects” endeavour
• Project Affiliation and Affiliated Projects - Ongoing reviews
• Miscellaneous HSF matters
  • HSF GitHub matters
• AOB
  • Next Meeting

Attending

Present/Contributing: Eduardo Rodrigues, Claire Antel, Graeme Stewart, Stefan Roiser, Johannes Elmsheuser, Nicole Skidmore, Liz Sexton-Kennedy, Michel Hernandez Villanueva

Apologies: Caterina Doglioni

News

Joint WLCG/HSF workshop 2026 in Bologna

• Indico: https://indico.cern.ch/event/1695268/.
• Event went online with registration and call for abstracts for the HSF parallel sessions.
  • No flood of registrations so far but expected as we’re in holiday season.
  • Hoping for a good representation from us!
• Set of HSF parallel sessions drafted. Will be populated and adapted as contributions come in.
• Do you have suggestions for speakers we should poke/invite? Please everyone have a thought.

AG meeting at the workshop

FYI we’re trying to arrange another AG meeting during the Bologna workshop to (1) avoid having a long 1-year delay (as in the past) while (2) profiting from the fact that most AG members will be at the workshop anyway.

Activity Area activities

Action from the last meeting, to liaise with the conveners to check status and plan of activities - DONE.

WLCG Management Board community software liaison

The next LHCC meeting will be in one month. HSF is NOT being asked for a report. Instead, the topical presentation will be on IML.

A WLCG Open Technical Forum (OTF) is planned for 25/26 Aug, https://indico.cern.ch/event/1683255/. The agenda will also contain a discussion on the status of WLCG heterogeneous computing infrastructure.

Advisory Group news and follow-ups

Actions from the last meeting:

• Attract representatives from ALICE and MCnet on the SG team - DONE for MCnet, no representative from ALICE yet found.
• Representation from ePIC/EIC on the AG - DONE.

“Responsible AI-assisted coding and common training aspects” endeavour

• Kick-off meeting held yesterday: https://indico.cern.ch/event/1705480/.
• Really well attended, peaking close to 200 participants (most people online).
• A lot of work to do to follow up.
  • After meeting follow-up email: share discussion summary and highlighted points, follow-up on sli.do questions.
  • Carefully plan next forum discussion, likely: More around open discussion, more focus on non-CERN experiments/labs, more focus on guidelines?
    • There are a few initiatives in Germany: DESY Assistant and Helmholtz AI (which are multi-science).
  • (Claire) Would also like to start something concrete (we can go on discussing these open issues forever), e.g. docs repo to collect AGENT.md, CONTRIBUTING.md, etc, files? SKILLS files would be useful to understand better.
• For training, following up with Graeme on LLMs running inside an isolated container.

Project Affiliation and Affiliated Projects

We had finalised a significant review of the programme but have not had any new affiliations since then.

ACTION: discuss the list of projects to prioritise to send to the AG chair (to broadcast to the experiments/collabs for feedback), and attach names to make things happen - basically a person to take care of finding 2 reviewers and potentially be a reviewer. - Typical pool of reviewers: AA conveners, SG members, HSF affiliated project leads, expert community colleagues.

List drew from recent brainstorming, including during coordination meetings:

• ACTS - Caterina to follow-up
• Clad
• CVMFS
• DD4hep
• Gaudi - Stefan to follow-up
• Geant4
• HepData
• HepMC3
• HERWIG
• hls4ml - Eduardo to follow-up
• Invenio
• Omnifold - Saptaparna to follow-up
• MCFM
• Phoenix - Johannes to follow-up
• PODIO - Michel V. to follow-up
• Rivet
• Rucio
• XRootD

ACTION FOR ALL: find one person to follow-up on each project.

Ongoing reviews

• NoPayloadDB Conditions Database - still WIP, there’s been quite some delay, unfortunately.

Miscellaneous HSF matters

HSF GitHub matters

Quick thoughts on #1919.

Reviewed problems of some recently closed PRs by non-HSF people:

• PRs implement unnecessary changes.
• PRs implement unasked for changes (like addition of tools or web function we never discussed we needed).
• PR flooding of little changes.
• Accidental AI contribution in one case? PR deleted some text and didn’t match description of PR!

Thoughts: I don’t think AI is directly our problem. I don’t think a pull request template asking for the amount of AI used is addressing any problems.

Plans:

Make a CONTRIBUTING.md for the HSF repo: Meant to document GitHub repo contribution guidelines - link to it automatically displayed when someone opens an issue/PR + linked under a “Contributing” tab on the repo. -> Makes any statement on our actual website unnecessary I think…

CONTRIBUTING.md content:

• Etiquette
• Due diligence
• Good practice/considerations
• Reasons PRs may be rejected.

Questions discussed:

• How useful is it to ask for indication if AI was used? We should be checking every PR regardless.
  • Other SG members find it useful.
• Is it useful to require an issue be made first if improvements to webpage/general update across all pages want to be made?
  • General agreement.
• Is it useful for us to encourage contributions to open issues, if they’ve been assigned to other HSF members?
  • Yes - even if assigned.

AOB

Next Meeting

TBD, as we see fit, but not before September.

---