DAAI · Document IntelligenceAI & AutomationCase study
Document AI Platform
Turns invoices, forms and IDs into clean structured data without anyone keying it in.
AI-powered document processing platform that automatically extracts, classifies and validates data from invoices, forms, IDs and PDFs using a multi-engine OCR + NLP pipeline running on async Celery workers.
Document AI Platform is an enterprise-grade solution that turns unstructured documents into clean, structured data. It automatically classifies document types and extracts fields from invoices, forms, identification documents and PDFs — at high accuracy and a fraction of the manual effort.
The processing core is an asynchronous pipeline: FastAPI accepts uploads and dispatches work to Celery workers (brokered by Redis) so large or scanned documents never block a request. Each document flows through a multi-engine OCR stage (PaddleOCR, Tesseract and OpenCV pre-processing), layout parsing, and an NLP/embedding stage, with Google Document AI / Vision available as a high-accuracy path.
A Next.js dashboard lets operators monitor pipelines, review extracted data and manage workflows, with audit trails for compliance. The whole system is containerized behind nginx for production deployment.
The problem
Enterprises spend countless hours manually keying data from invoices, forms and IDs — slow, error-prone and expensive, with large scanned PDFs making naive processing time out.
The solution
Built an async, multi-engine document pipeline: FastAPI dispatches uploads to Celery workers, which run OCR (PaddleOCR/Tesseract + OpenCV), classification and NLP extraction, falling back to Google Document AI for tough documents.
Technical stack
Frontend
Next.js
Backend
FastAPI
Python
Celery
Data
Redis
PostgreSQL
AI & Processing
PaddleOCR
Tesseract
OpenCV
Transformers
Google Document AI
Infrastructure
Docker
Nginx
System architecture
1
FastAPI API layer accepts uploads and enqueues jobs; Celery workers (Redis broker) run the heavy pipeline asynchronously with a scheduler for recurring tasks
2
OCR stage combines PaddleOCR and Tesseract with OpenCV pre-processing; category templates drive classification and field mapping
Google Document AI / Vision available as a high-accuracy path; cloud storage via S3 (boto3)
5
Next.js dashboard for monitoring, review and workflow management; PostgreSQL for data and audit trails
6
Dockerized and served behind nginx for production
Engineering challenges
Challenge
Large and scanned PDFs blocked synchronous requests and timed out.
How I solved it
Offloaded processing to Celery workers with a Redis broker, so the API returns immediately and work runs asynchronously.
Challenge
No single OCR engine handled every document type well.
How I solved it
Combined PaddleOCR, Tesseract and Google Document AI behind template-based classification to route each document to the best extractor.
Core features
Automatic document classification with category templates
Multi-engine OCR (PaddleOCR + Tesseract) with OpenCV pre-processing
NLP & embedding-based field parsing
Invoice, form and ID document extraction
Async processing on Celery workers — no blocked requests
Real-time processing dashboard
Batch processing support
API-first integration
Audit trail & compliance
High accuracy (95%+)
Results & impact
Reduced document processing time by ~85% and lifted data accuracy to 95%+, turning a manual back-office bottleneck into an automated, auditable pipeline.
Screenshots
What I learned
Async pipelines (FastAPI + Celery) are the right shape for unpredictable, heavy document workloads.
Multi-engine OCR with template-driven routing beats betting on a single model.
