Name is Yuiry - a software engineer focused on building self-hosted AI systems — local LLMs, AI agents, and retrieval-augmented generation (RAG) pipelines.
Right now, my work is centered on:
- 🤖 LLM agents with tool use
- 📁 RAG apps with self-hosted models
- 🧠 Autonomous reasoning systems using local infrastructure
-
🔌 Your Very Own Chatbot
Local AI assistant using Ollama -
🧠 Airwindows Expert
RAG-based chatbot with FAISS-powered document search, grounding on Ollama + LangChain-style stack. -
⚙ GitHub Issues MCP Server
Self-hosted, stdio-powered MCP Server to interract with GitHub Issues.
Core Tools:
Ollama • FAISS • LangChain • Python • FastAPI • Docker • Git
AI Focus Areas:
- 🧠 Local LLMs (LLaMA, Mistral, Code Llama)
- 🛠️ Tool use (commands, context switching, file operations)
- 📚 Retrieval-Augmented Generation (custom datasets)
- 🧩 Agent workflows (planning, chaining, decision logic)
- 💼 Not long ago PHP backend developer (Symfony, REST APIs, microservices)
- 🔄 Transitioning into AI engineering — bringing experience from backend design into intelligent systems
- 🧱 Strong grounding in system design, data flow, and integration
- Expanding agent frameworks using local LLMs + toolchains
- Testing open-source inference backends (llama.cpp, vLLM, Ollama)
- Experimenting with LangGraph, CrewAI, and advanced orchestration
- 🧠 Open to AI agent prototyping, OSS collaboration, or early-stage AI startups
- 📬 Reach me via GitHub or Upwork
"Make AI useful, not just impressive."