Part 3 · CogniVault Architecture: Why We Keep Ollama Out of Docker
Containerizing an entire AI pipeline sounds great until the GPU gets involved. Here is how CogniVault safely bridges Docker containers with native host execution.
Part 1 · Why I Built a Local-First RAG
Cloud AI assistants are powerful — but for trainers, researchers, and anyone handling sensitive material, they're also a leaky abstraction. Here's why I built a 100% local …