Date: June 16, 2026
Amit Bashyal, Kolja Kauder: BNL, Derek: JLab, Anselm Vossen: Duke, Hemalata, James, Cristiano: W&M
DOE – progress report approved by DOE
Creation of CUA as vendor — Cris -> reach to Tanja
Shifting the technical meeting of a week
Derek presented several PRs (one fixing a few bugs causing pytests to fail for the optimizers, configurations, scheduler, workflows modules; two other PRs working through the remaining bugs blocking examples/epic/epic_stack_example.py to have a complete, working from-YAML-only example)
Derek and Cynthia - found out that CLI is not fully implemented yet, work in progress
Cynthia is doing work on the ePIC build (at every trial; not doing that would be faster) as well as how to submit sub-jobs; Wen reminded that sub-jobs were handled when working with PanDA at the beginning. Amit will also look at multi-stage PanDA - split one job in multiple sub-jobs and run it
We agreed on the fact that for the paper we need one working pipeline for one subdetector, supporting https://github.com/aid2e/AID2E-framework/; for stress-testing complexity of larger detector design in principle we can duplicate that pipeline. If all sub-detector pipelines are ready to be merged, then we will use a unified detector design for the stress-tests.
We identified a list of tasks
1) (Amit + Derek) CLI(s) needed for DTLZ tests with different schedulers and optimizers 2) (Cynthia, anybody else?) no way to collect objectives (missing utility to map that output back to Ax/optimizer) - [drich_utils.py: collect_objectives, failed_trials_from_stage_result] 3) (Cynthia) try to figure out best way to implement per trial subjob support → merge in AID2E framework [drich_trials.py] 4) (Cynthia) PR for review after PR#53 merge for adding the no-per-trial build geometry option [drich_utils.py, prepare_trial_geometry] 5) Work on multi-stage (different schedulers) to support EIC workflows (Derek, Amit, Cynthia — see number 2) 6) Converging on one single sub-detector workflow (either dRICH, BIC, far-forward) and potentially duplicate that for stress-tests — specific to subdetector workflows (Baptiste (FF), Cynthia (dRICH), Derek (BIC) + SLURM/PanDA (BNL, W&M)). Notice:
- (Derek) The workflow introduced in PR#53 is sufficient for the paper: it’s a SOBO on the BIC AstroPix staves using the photon phi resolution as an objective at 3 kinematic points. I’ve tested the analysis script and kinematic points in the “standalone” (until we get the FW running) BIC-MOBO repo, and Amit and I have confirmed it runs as expected with Ax + SLURM/PanDA
- (Baptiste, Cynthia) checking working of FF and dRICH pipeline with PanDA scheduler
7) Cris + anybody else? — working on the Overleaf skeleton
The majority of time at the past technical meeting was dedicated to discussion of Phase I. We said to Amit (presenting updates on Phase II) to share more details and provide an update during this General Meeting if time permits. – Agreed?
Some notes taken last time: Testing the preliminary AID2E agent (June 11 slides)
Test done with preliminary AID2E agent
Next technical meeting - Roadmap, distribution of tasks for Phase II