# AiiDA Team Meeting 2024-08-22

###### tags: `team meetings`
###### time: 14:00 CET

[TOC]

### Present

* Julian
* Marnik
* Miki
* Xing
* Julian
* Sebastiaan
* Alex
* Ali

### Catch-up round

*Max. 3 minutes each*

Julian:

* AiiDA talk in former group in Spain (25/07), and at New Mexico State University (NMSU, 09/08)
* AiiDA talk (20min), hands-on tutorial (2h), and panel discussion (90min) representing NM and GP :D at [FAIR workflows workshop](https://nanohub.org/FAIR_workshop_2024) at Purdue University (13-14/08). Recordings should be online (YouTube?) soon.
* Currently working on [90min tutorial for EuroSciPy](pretalx.com/euroscipy-2024/talk/BYESWT/) (27/09) with Alex and Ali showcasing aiida-shell and aiida-workgraph
* LinkedIn profile for AiiDA team, and AiiDA page (see below for discussion)
* Working on group/profile dumping feature, dogfooding welcome soon :pray:

Marnik:

* Nothing to report on the AiiDA-core front.
* Pinning down dates for AiiDA retreat: 18-22 Nov: for whom does this _not_ work?
* Wrote to [Central Residence in Leysin](https://hotelcr.ch/home) to ask for budget.
* Asked Viviana for other possible locations (more nearby PSI for the poor people living in AG.).

Sebastiaan:
* Invititation to speak at a conference for materials science in Julia community organized by [Carnegie Mellon](https://acme-group-cmu.github.io/). Cannot make it myself. Anyone else interested? They would accept a remote contribution if necessary.
* Will work on [ORM: Use pydantic to specify a schema for each ORM entity](https://github.com/aiidateam/aiida-core/pull/6255). Will update the [associated AEP](https://github.com/aiidateam/AEP/blob/983a645c9285ba65c7cf07fe6064c23e7e994c06/010_orm_schema/readme.md) with the current state.

Edan:
* Finally got around to [Seb's Pydantic models PR](https://github.com/aiidateam/aiida-core/pull/6255). Will be testing locally for application in the REST API for the query builder webapp/widget

Ali:
* [robots.txt](https://github.com/aiidateam/aiida-core/issues/6516) not in the right place, google returns outdated results.
* [Stashing](https://aiida.readthedocs.io/projects/aiida-core/en/stable/topics/calculations/usage.html#stashing-on-the-remote) How do you unstash?
* Need feedback on [Add filters to verdi group delete](https://github.com/aiidateam/aiida-core/pull/6556)

Xing:
* AiiDA-WorkGraph projects
* [For stable release](https://github.com/orgs/aiidateam/projects/9/views/1)
- Add [While task](https://aiida-workgraph.readthedocs.io/en/latest/howto/while.html)
- Add [If task](https://aiida-workgraph.readthedocs.io/en/latest/howto/if.html)
- Improve the syntax of
- [waiting_on](https://github.com/aiidateam/aiida-workgraph/pull/223): Although there remains some debate over whether to use the `append` or `replace` method for managing the `waiting_on` list, I've decided to proceed with the append approach.
- Use entry point to register task
- Add [monitor task](https://aiida-workgraph.readthedocs.io/en/latest/howto/monitor.html): This is useful for scenarios, including time-based triggers, file existence checks, and monitoring other tasks or WorkGraphs.
* [Next release](https://github.com/orgs/aiidateam/projects/10)
* [Add awatiable_builder decorator](https://github.com/aiidateam/aiida-workgraph/pull/239): The primary purpose is to allow the WorkGraph to submit other jobs. E.g, Implement the `submission_controller` using WorkGraph.

Miki
* StructureData is here: I put a message on [discourse](https://aiida.discourse.group/t/new-structuredata/451).
* Working with Yuhao for magnetic calculations in aiida-quantumespresso.
* to be done: small decision on naming some methods (see on [discourse](https://aiida.discourse.group/t/new-structuredata/451/2?u=mikibonacci) the comment from Gio)

Alex:
* [AiiDA-WorkGraph]
- Initialize frontend test infrastructure with rudimentary tests
- Initialize doc tests using sphinx-gallery to run examples in the docs

Jason:
* NUMFOCUS newsletter for next month. (hero run and v2.6 release.)
* Finalize aiida-hyperqueue, it is now used in QeApp demo server as the lightweight scheduler. But it does not have concept of `num_machines`, I put `num_cpus` and `memory_mb` as the field which mirror the concept from hyperqueue.
* The limit of 200 slots per worker is blocking me to run a lot SSSP verifications (aiida-submission-controller already helps a lot.). Any foreseeable problem if I increase the worker slots from 200 to 20000?

### General discussion

* Julian
* Created an AiiDA team profile on LinkedIn using the aiidateam Gmail address: https://www.linkedin.com/in/aiida-team-1190b5324/ <- Please all connect!
* PSI as institution? Or EPFL? I think now should be PSI.
* Verification requires PSI-mail, which wasn't used for LinkedIn profile verification already (I already used mine for that)
* [GP] let's use info AT aiida.net
* With this user, we want to create a Business page (educational institution):
* URL: `linkedin.com/school/aiida` already taken
* [GP] which URL
* Suggestions:
* aiidateam
* aiida-python
* the-real-aiida
* Regular posts? Every Monday, every Friday? Weekly, bi-weekly, etc.
* Start with announcement of EuroSciPy talk and tutorial
* Then, blog post with features of v2.6 release
* Next -> @Seb, you had some blog posts ready, as well?
* Planning document: https://hackmd.io/_Ic1IR6pQ6yZYnqZUflMjg