When you get enough complexity, predictability goes down. LLMs are showing emergent behavior difficult to gauge from dismissing them as being autocompletes on steroids.
What follows is an actual session I just had with the LLM I’m using for hobbyist development, Devstral-2, using the Mistral Vibe CLI. The application is a mod I’ve developed for the game Hytale .
AI · Development · LLM · Mistral · Vibe
1019 words
5 minutes
How to use Mistral Vibe as a development companion that remembers the context of the different projects you work on
I’m using Mistral Vibe as my “AI CLI” when doing LLM-aided development. Sometimes I use local models, sometimes their cloud models, it depends on the sensitivity of what I’m working on. I grew a bit tired of having to repeat what I was working on when picking up each project again though and this post describes what I did to get a more pleasant LLM-experience.
AI · Development · LLM · Mistral · Vibe
368 words
2 minutes