Service

Document Intelligence

OCR plus LLM extraction pipelines that turn documents into structured data.

What I build

Pipelines that take messy documents in and give typed JSON out: invoices, contracts, ESG reports, forms. OCR handles the scans, an LLM fills a strict schema, and an async worker layer keeps the API responsive while the slow work happens.

How I work

Schemas first. We define the fields, their types and their validation rules, then measure extraction accuracy on a labelled sample before any tuning. New document types become a template, not a project.

When to call me

A team is retyping documents into a system by hand, or an existing OCR setup gives you text but not data.

What you get

Extraction API with async processing and status tracking
OCR and layout handling for scans, PDFs and images
Typed output schemas per document type, with templates for new ones
Local or cloud LLM backend, switchable by configuration

Selected work

Case studies for this service

Self-Built Project

Demonstrated

Document Extraction Pipeline: Scaling AI Workloads with FastAPI and Celery

An async-first extraction service: FastAPI accepts uploads, Celery workers run OCR and LLM extraction, and typed schemas turn invoices, legal documents and ESG reports into JSON.

Demonstrated capabilities

These are implemented capabilities and supported schemas; no comparative speed or accuracy benchmark is claimed.

Async

Queued worker processing keeps uploads responsive

Typed JSON

Schema-validated output for supported templates

Read case study

Related work

Related projects

Document Extraction Pipeline

Turn Documents into Structured Data with AI

Self-Built ProjectDemonstrated

A production-ready FastAPI service that extracts structured data from PDFs and images using OCR and LLM technology. Features async processing, JWT authentication, and support for multiple document types including invoices, legal documents, and ESG reports.

FastAPICeleryPostgreSQL

Questions I get asked

What accuracy can I expect?

On structured documents such as invoices and forms, above 95 percent field-level accuracy after schema tuning, measured on your own samples. Free-text legal documents need a review step and I design for that.

Can it run without sending documents to a cloud LLM?

Yes. The pipeline swaps between a cloud model and a local one served with Ollama without code changes, which matters for contracts and personal data.

How does it scale?

The API accepts uploads and returns immediately; workers process documents in the background. Throughput grows by adding workers, not by rewriting anything.

Have a problem that looks like this?

Tell me about it. I reply within one working day with a first take and no sales pitch.

Book a call Send a message