News for hackers. Stuff that matters.
Re:Pandas 3.0 by optimalsolver on Wednesday January 28, 2026 @05:58AM (#46793751)
How soon will the leading LLMs ingest the updated documentation? Because I'm certainly not going to.
Re:Pandas 3.0 by uncletoxa on Wednesday January 28, 2026 @06:16AM (#46793885)
Use context7 mcp. It'll do the trick
Re:Pandas 3.0 by leadingthenet on Wednesday January 28, 2026 @10:40AM (#46796776)
I've been sleeping on this, works like a charm!
Re:Pandas 3.0 by OutOfHere on Wednesday January 28, 2026 @06:30AM (#46793987)
In my experience, it would take a year to ingest it natively, and two years to also ingest enough coding examples.
Re:Pandas 3.0 by swyx on Wednesday January 28, 2026 @12:27PM (#46798550)
its not even about the ingest, every major semver change now is a problem because now LLMs will need to contextually distinguish whether or not they are expected to output Pandas 2 or 3, unless ofc you explicitly prompt it.
Re:Pandas 3.0 by OutOfHere on Wednesday January 28, 2026 @01:01PM (#46799123)
I wouldn't worry about it because over a longer period, this automatically leans toward the more recent versions. There are multiple forces that exist to make this happen.

The main exception is for legacy code requiring maintenance when they are unwilling to upgrade Pandas.

Re:Pandas 3.0 by swyx on Wednesday January 28, 2026 @01:15PM (#46799358)
yes but a lot of legacy code wil still need to be maintained and written. you dont see how this can be confusing/annoying?
Re:Pandas 3.0 by g-mork on Wednesday January 28, 2026 @09:30AM (#46795895)
This is the most misunderstood aspect of how marketing has changed recently
Re:Pandas 3.0 by esafak on Wednesday January 28, 2026 @11:24AM (#46797459)
You could create skills out of the docs if you use it a lot. https://agentskills.io/