Challenge
Ocean-carrier service contracts vary widely in structure, terminology, and table layout. Teams were spending hours manually locating and encoding critical terms while maintaining little traceability back to the source document.
Manifesto: Freight Contract Extraction Pipeline
The project
An automated and traceable freight-contract extraction workflow designed to reduce manual contract data entry from hours to minutes.
Technology
Python · FastAPI · Next.js · React · TypeScript · Tailwind CSS · PostgreSQL · Supabase · pdfplumber
Approach
We designed a guided extraction pipeline that combines document parsing, schema-aware review, page-level source references, and a human verification workspace. The system supports AI-assisted unknown contracts and more deterministic known profiles.
Outcome
Manifesto won OLTEK’s From Paper to Data challenge and continued into product refinement. The workflow reduces repetitive encoding while keeping every extracted value reviewable and traceable.
Contribution
I led product direction and presentations, translated operational requirements into workflows, and contributed across the Next.js and FastAPI implementation while coordinating delivery with the team and OLTEK.
