01 / Projects
AI Yacht Pricing System
A technical-data system that turns yacht documentation into structured knowledge and pricing logic.
The project is shown at the level of problem, architecture and capabilities; some implementation details remain reserved.
01 · The problem
Technical documents, PDFs and scattered specifications cannot be priced by hand without losing time, consistency and traceability.
02 · The idea
Treat pricing as a data problem first. Extract, normalise, store, then apply AI only where judgment is needed.
03 · The system
A pipeline from technical documents to quotation: extraction, normalisation, database, AI assistance, pricing logic.
04 · The implementation
Built around a PDF intelligence pipeline, a structured database and an AI layer that reasons over normalised technical data — not over raw files.
05 · The next step
Complete the quotation loop, tighten validation and connect the knowledge engine to the engineering assistant.
The architecture
- 01 Technical documents
- 02 PDF extraction
- 03 Normalisation
- 04 Database
- 05 AI
- 06 Pricing logic
- 07 Quotation
Technologies
- Python
- FastAPI
- PostgreSQL
- SQLAlchemy
- vLLM
- Qwen
Categories
AI · Engineering · Data