Josh Temple (@joshptemple) 's Twitter Profile
Josh Temple

@joshptemple

I build data platforms and make ice cream. Co-founder of @SpectaclesCI. Previously @Spotify, @milkbarstore.

ID: 3029358550

linkhttp://joshtemple.dev calendar_today11-02-2015 02:39:34

418 Tweet

395 Followers

422 Following

Josh Temple (@joshptemple) 's Twitter Profile Photo

Today I discovered the Python package "polyfactory" by Litestar , which uses Faker and Pydantic to generate believable mock test data. It's a combo that just makes sense.

Max (@max_sixty) 's Twitter Profile Photo

Charlie Marsh Would be keen to see an experiment abandoning traditional bug reports and instead requiring a PR containing an xfailing test Then the fix just has to pass that test I've thought about enacting it, but it seems a bit too difficult for newbies / unfriendly

Bartłomiej Cupiał (@cupiabart) 's Twitter Profile Photo

So here's a story of, by far, the weirdest bug I've encountered in my CS career. Along with Maciej Wołczyk we've been training a neural network that learns how to play NetHack, an old roguelike game, that looks like in the screenshot. Recenlty, something unexpected happened.

So here's a story of, by far, the weirdest bug I've encountered in my CS career.

Along with <a href="/maciejwolczyk/">Maciej Wołczyk</a> we've been training a neural network that learns how to play NetHack, an old roguelike game, that looks like in the screenshot. Recenlty, something unexpected happened.
Learn Prompting (@learnprompting) 's Twitter Profile Photo

🚨 Role Prompting doesn't work... Our team at Learn Prompting led a year-long study with co-authors from OpenAI & Microsoft, analyzing over 1,500 prompting papers. We narrowed it down to 58 different prompting techniques and we analyzed every one. Here's what we found...

Amanda Mull (@amandamull) 's Twitter Profile Photo

In my first feature for Businessweek Bloomberg, I dug into a question that everyone I know has had for years: Putting everyday goods like toothbrushes and deodorant behind locked plexiglass can't actually be working out well for the stores that do it, right? bloomberg.com/news/features/…

Josh Temple (@joshptemple) 's Twitter Profile Photo

If you’re considering using LLMs for hiring decisions, think again. “In all three MTE models, white names were preferred in a full 85.1 percent of the conducted tests, compared to Black names being preferred in just 8.6 percent.” arstechnica.com/ai/2024/11/stu…