Marcin Mazurek
Product · Technology · Curiosity
Projects
-
Local Sovereign AI #1 — Running Local LLMs →
How to run open-source LLMs locally with Ollama — privacy, cost, and control.
-
Local Sovereign AI #2 — LiteLLM as Your Local AI Gateway →
A unified gateway in front of all your models — local and cloud — with routing, cost tracking, and fallback logic.
-
Local Sovereign AI #3 — RAG on Your Own Data →
Retrieval Augmented Generation running entirely locally — query your documents without sending anything to the cloud.
-
Local Sovereign AI #4 — MCP Protocol Explained →
The open standard that gives AI agents controlled, auditable access to live data sources — databases, APIs, file systems — without bespoke integrations.
-
Local Sovereign AI #5 — Running Open Source Models at Scale →
vLLM vs. Ollama vs. LM Studio — when Ollama isn't enough and what to reach for when you have real throughput requirements.
-
Local Sovereign AI #6 — Local AI Observability →
Tracing, logging, and monitoring for local LLMs — because AI systems fail silently in ways traditional monitoring completely misses.
-
Pyramid builder →
Pyramid builder.