logo FEDIDEVS

DjangoCon US 2026

Benedikt (12)

Django ORM Kung Fu - Postgtes can do a lot more in the ORM than you might think and being a Bass player in a Jazz band might help understand those techniques better. At least Mathias Wedeken said in his talk at
Tricks used: Composing annotations like functions, Func(template=), ...

0 1 0

For anyone not in the Chat - you should absolutely check out the blog by Reinout van Rees - reinout.vanrees.org/weblog/ind

He's doing really comprehensive summaries of all the talks he's visited and keeps documenting them for posterity.

And plugging his vacation photos 😅

1 0 0

Very interesting talk about packaging and shipping your Django App to run in a Desktop Context through Electron by @jochen - It's apparently not that different from running on the web and since there's only one user at a time, it needs way less resources.

#electron

4 3 0

In the 2nd Talk of the day - Andrew Northall taked about Caving (he really does that for fun!?) and how he collects and analyzes accident reports into structured data. One problem was the specificity of Data - he solved that by adding Treebeard to have a hierarchy of data that can be searched.

0 1 1

Kickoff for Thursday at #Europe. 🇬🇷 - @Marlene gives a beginner friendly Intro to Agentic Coding and shows how to have AI generate your Playwright Tests or Diagrams in Excakidraw. For advanced users the small example using the Copilot SDK should be very interesting I think.

0 0 0

Very advanced database stuff for very big tables - @timb07 talked about database partitioning and especially applying partitioning to an already existing database. The screenshot shows one of the queries needed to perform the migration.

5 3 1

Skipped the afternoon talks to attend a Workshop with @thibaudcolas
He talks about llms.txt as a standardised format for writing documentation optimized for LLM agents showing both the good (lighter docs, easy to find, ...) and bad (prompt injection, no ad traffic, ...)

0 0 0

Just finished a great talk by Haki Benita where he proposed reliable signals - basically a signal whose only job is triggering a Task.
The example used in the talk (a payment flow) is pretty much the textbook example of why the Django Docs warn to not mistake signals for loose coupling .

3 2 1

Another talk about what AI does to Open Source projects (there will be more positive AI talks later in the conference). Here @paulox shows examples of how maintainers feel about AI and what rules they set in their projects.

4 1 1