RestoreRAG
A chatbot that answers questions from restoration industry documents.
- Searches documents by meaning, not just keywords
- Answers are based on the source files
GitHub
Live demo coming soon
I build AI that answers from your documents.
Senior software engineer. I build RAG chatbots, document extraction and AI automation for businesses, using Python, FastAPI and Gemini.
How a document chatbot works
1. Upload documentsPDFs, manuals, reports, scans
2. Index by meaningEmbeddings stored in a vector database
3. Ask a questionThe most relevant passages are retrieved
4. Get a grounded answerAnswers come from your files, not guesses
Selected projects
Two working AI systems. Live demos and write-ups are being added as each one is polished.
DocuExtract
Turns documents in many formats into clean, structured data.
- Reads PDFs, images and scans with vision AI
- Outputs fields you can send straight to a database or sheet
GitHub
Live demo coming soon
What I can build for you
Small, focused projects with clear deliverables.
RAG chatbots
Ask questions across your manuals, policies or reports and get sourced answers.
Document extraction
Invoices, receipts and forms converted into structured data automatically.
AI automation and APIs
Python and FastAPI backends that connect AI to your existing tools.
About me
I am a senior software engineer at a government water utility, where I build mobile apps, backend systems and geospatial tools used by real staff and citizens. I hold an MS in Computer Science from NED University.
I now apply that production experience to AI projects: practical systems that work reliably, not demos that break.
Tech I use
PythonFastAPIFlaskGemini APIRAGpgvectorSupabaseFlutterFirebaseMySQLDockerQGIS
Have a project in mind?
Tell me what you need. I reply with a clear plan and timeline.