#BuildPropulsionLab at #DevoxxUK: @jbaruch with @martinfoertsch, @mayerjrrr and @originalOne1984
summary
Consultancy teams report using Llama 3s and Mistral models deployed in their own data center, fine tuning for legal or industry jargon, and writing custom VS Code/JetBrains plugins and a private chat front end to avoid third‑party data leakage. They claim these open‑source stacks get development and migration tasks 80–90% of the way, speed up debugging and boilerplate generation, and note that bespoke plugins plus fine‑tuning are required for the rest.
AI-written summary · about
related talks
12mAI-generated recap of Devoxx Belgium 2024 opening keynote : Genesis
58mGenesis Keynote
30mDevoxxGenie: Your AI Assistant for IDEA
46mDeveloper Productivity Engineering at CERN
27mQ&A #2
50mDeveloper Productivity – DIY (with LLM) or How I Learned to Stop Worrying and Love the AI
25mSelecting the Best LLMs with Model Cockpit
37mEmpowering Developers with local AI