Back to feed
Dev.to
Dev.to
7/4/2026
Building an Offline Document Search Engine for My University

Building an Offline Document Search Engine for My University

Short summary

A Turkish university student built an offline document search chatbot that scans regulation PDFs to answer questions deterministically without hallucinations. Uses Python/FastAPI with BM25 search optimized for Turkish language and custom PDF parsing. Open-source and deployable locally—useful as a template for document-search systems in any domain.

  • •Offline-first document search for university regulations, avoiding cloud LLM hallucinations
  • •Implements custom BM25 search with Turkish-language tokenization and synonym expansion
  • •Open-source, deployable locally via Docker; extensible template for document-search applications

Generated with AI, which can make mistakes.

#ai-tools#open-source#ai-agents#certification-education
Read full article at Dev.to

Is this a good recommendation for you?

Comments

Failed to load comments. Please try again.

Explore more

How I Built an Ultra-Fast, Programmatic Results & GPA Portal for My University (MUET)
Dev.to
The original title is "Cortexa: a free open-access search engine for ML and robotics papers"
Dev.to
I Built a Local RAG Assistant with Ollama, ChromaDB and LangChain. Here's What I Learned
Dev.to
Building a RAG Chatbot with FastAPI and ChromaDB (that runs locally, no API key)
Dev.to