I will build a private local rag system with ollama and fastapi

D
dadashovvvh
D
dadashovvvh
Tural Dadashov

About this gig

I will build a fully private, on-premise RAG (Retrieval-Augmented Generation) system using Ollama, Qwen, FastAPI, and Docker. Your documents never leave your computer 100% local, zero data leaks.


WHAT YOU GET:

Custom AI assistant that answers questions from your PDFs, Word, and Excel files

Exact source citations no hallucinations, every answer references the document

Local deployment no cloud, no OpenAI, no third-party access

Secure API access via Cloudflare Tunnel

Docker containerization for easy deployment

Perfect for legal, finance, healthcare, and government sectors


TECH STACK:

Ollama + Qwen 2.5 7B (16k context)

FastAPI backend

Vector database (Qdrant / pgvector)

Docker


WHY ME?

6 months of Web Pentest (cybersecurity) experience security-first approach

Hands-on experience with local LLM deployment

Live demo available test before you buy


HOW IT WORKS:

1. You send me 3-5 sample documents (PDF, Word, Excel)

2. I set up the entire system on your machine

3. I give you API access and documentation

4. You test it, I provide 3-7 days support


️ Delivery: 3-5 days

Support: Included


Let's build your private AI assistant today!

Get to know Tural Dadashov

Tural Dadashov

AI Engineer and RAG Specialist

  • FromAzerbaijan
  • Member sinceSep 2026
  • Languages

    Azerbaijani, English, Turkish
I specialize in building private, on-premise AI systems using Ollama, Qwen, and FastAPI. My solutions keep your data 100% local – no cloud, no data leaks. I have hands-on experience with RAG pipelines, vector databases, and Docker deployment. Live demo available.

My Portfolio