Exploring AI that respects your data.
This site covers artificial intelligence in practical terms — what’s actually being built, what it’s useful for, and what it means for people who care about where their data goes. The short version: practical AI, privacy first.
Beyond the blog, the plan is to grow this into a broader resource — tools, platform guides, and reference material for anyone building or using AI outside the mainstream cloud ecosystem.
Our platform and all our services are non-commercial for personal use only. Our services are currently available only for contributors and friends.
What you’ll find here
Four topics come up repeatedly, because they reflect what’s actually being worked on and thought about:
- AI tools & models — new releases, real-world capability assessments, and what actually changes when a new model drops
- Digital privacy — as AI systems get more capable and more connected, the question of what data goes where matters more than ever
- Platform updates — what’s being built and deployed on the self-hosted stack behind this site
- Resources & guides — practical reference material for running capable AI on your own terms
The stack
The platform behind bacher-ai.com runs on self-hosted and carefully vetted infrastructure. Where third-party services are used, they are selected for confirmed privacy practices — no training on user data, and no data retention beyond operational needs.
- Open WebUI — the interface for interacting with local and remote AI models
- Ollama — runs open-weight models locally, keeping inference fully on-device
- LiteLLM — a unified API layer routing requests across models while keeping control local
- Venice.ai — a privacy-respecting inference API with a confirmed no-training, no-retention policy
- OpenRouter — access to a broad range of models via a single API, used selectively for specific capabilities
- n8n — workflow automation that connects tools, triggers actions, and powers much of what gets published here
Posts here often reflect what’s being learned on that platform directly — which means the content is grounded in actual use, not speculation.
Who this is for
If you’re interested in capable AI without compromising on privacy, you’re in the right place. The writing assumes a basic technical background — terms like API, LLM, and encryption won’t be explained from scratch — but the goal is always clarity over jargon. If you wish to sign-up for any of our services or want to contribute to the project, please get in contact with us.
Quote