Skip to Content.
Sympa Menu

phys-npps-members-l - Re: [[Phys-npps-members-l] ] AI-focused NPPS meeting

phys-npps-members-l AT lists.bnl.gov

Subject: ALL NPPS Members

List archive

Chronological Thread  
  • From: Torre Wenaus <wenaus AT gmail.com>
  • To: NPPS members <Phys-npps-members-l AT lists.bnl.gov>, "Qian, Xin" <xqian AT bnl.gov>, "Yu, Haiwang" <hyu AT bnl.gov>
  • Subject: Re: [[Phys-npps-members-l] ] AI-focused NPPS meeting
  • Date: Fri, 20 Jun 2025 11:15:41 -0400

An AI day update. I'm happy to report we'll have a second special guest from EDG, Haiwang Yu. 
July 2 is feeling too soon (and too close to the holiday), but perhaps we have too much agenda for one meeting and we should plan two, in which case July 2 may look OK for a first subset of topics.
Input welcome and we'll discuss the planning in our June 25 meeting.
Agenda suggestions/contributions are still welcome and expected!
  Torre

On Thu, Jun 12, 2025 at 9:29 AM Torre Wenaus <wenaus AT gmail.com> wrote:

Hi all,

I’ve started to draft an agenda for an AI-focused NPPS meeting. We’ll have Xin Qian from EDG as a special guest. The draft copied below is up in the meeting notes, comments and suggestions there or in this thread are appreciated. It includes draft slides, for developing before the meeting and discussing during it, on Supporting our Physics programs and developers with AI tools.


We have to decide when to have it. July 2 may be the only realistic date prior to August.

  • AI day - an NPPS meeting dedicated to AI/ML (mostly AI) topics. Draft agenda in progress, input appreciated:

  • when to have it - July 23? No, that conflicts with the REDWOOD workshop. July 2? Otherwise it might be August.

  • This meeting covers some new applications but the long agenda as a whole is less about applications and more about the tools - what are we using, what do we need, in the context of the applications we develop. What tools and techniques do you see as important for the work you're doing now and the future work you're still conceptualizing, and what should we and the lab be supporting to enable your vision. So application talks should be brief on describing the app itself (we can go into more detail in other meetings) and spend time also on the tools and techniques employed and how we should support them.

  • AI app experience & activities

  • PanDA MCP service proof of concept - Tadashi

  • ask-panda MCP service - Paul

  • TBD - Tasnuva

  • MCP in the ePIC streaming workflow testbed - Torre

  • Rucio MCP plans - Cedric?

AI tools and development assistants

  • Tools experience and Wire-Cell applications - Xin Qian

  • Experience with BNL’s Azure AI subscription

Tools & assistants update - Shuwei

Prompt engineering - tips and information sources - all

What lab support do we need? What should we ask for?

  • Round table on experience and wish list

  • Converge on a slide deck with the message to the bosses

Updating the BNL HEP AI/ML vision (2021)

How to keep mutually informed?

  • A standing ‘AI/ML tips’ item on the NPPS meeting agenda?

  • A collectively developed ‘tool tips’ googledoc? (the record of the standing agenda item)

  • A NotebookLM equipped with the context of what we do, charged to keep an eye on tools for us?

  • Joint Physics AI/ML meetings lapsed (my fault) since AI/ML became part of almost everything - should we have AI/ML focused meetings in Physics? Keeping in mind it’s a tool and not an end in itself (for us). 


  Torre


--
-- Torre Wenaus, BNL NPPS Group Leader, ATLAS and ePIC experiments
-- BNL 510A 1-222 | 631-681-7892


--
-- Torre Wenaus, BNL NPPS Group Leader, ATLAS and ePIC experiments
-- BNL 510A 1-222 | 631-681-7892



Archive powered by MHonArc 2.6.24.

Top of Page