I mean, can't you just… prompt engineer your way out of this? A writer friend of mine literally just vibes with the model differently and gets genuinely interesting output.
"Soon, Mariona joined her new friends on "raids": a few of them would block off a street, throw Molotov cocktails, hand out leaflets, and when the police turned up, scatter in every direction."
okay she threw molotov cocktails, she was lucky she wasn't imprisoned.
I'm utterly shocked at the article saying GPU inference (PyTorch/Transformers)isn't working. Numerical instability produces bad outputs,
Not viable for real-time serving, Wait for driver/CUDA updates!
My job just got me and our entire team a DGX spark.
I'm impressed at the ease of use for ollama models I couldn't run on my laptop.
gpt-oss:120b is shockingly better than what I thought it would be from running the 20b model on my laptop.
The DGX has changed my mind about the future being small specialized models.
I like the ORM but Django has stagnated in so so many ways.
Most of my startup friends basically use Ruby on Rails for their startup webap, and python microservices these days.
If you know python (hate ruby) and like javascript well enough FastAPI and javascript frontends seems way better.
TDD is a great example of where major differences between businesses and departments has direct impact on your software engineering.
When business people don't know what they want, do not try TDD. It will be a waste of time. When people do KNOW, or you have a RELIABLE subject matter expert (at a big company you might have one of these), TDD is a lot safer and easier to do.