AI-Powered Attendance Intelligence: OCR with Vision LLMs

Schools manually transcribe attendance sheets every day. It’s slow, error-prone, and consumes hours of administrative work. Traditional OCR tools struggle with handwritten or varied-format sheets. Enter Presença OCR.

The Solution

A Python/Flask backend using GLM-4 Vision LLM (ZhipuAI) processes scanned attendance sheets. Teachers upload photos and get structured data instantly — no manual transcription needed.

Architecture

• OCR Backend: Flask API with GLM-4 Vision for document understanding
• Frontend: TanStack Start with upload, review, and dashboard flows
• Storage: PostgreSQL with user management and role-based access
• Analytics: Attendance dashboards with Recharts visualization

Results

Zero transcription errors, 90% faster data processing. Teachers upload photos of attendance sheets and get structured data instantly, with an admin dashboard for analytics and export.

Built with: TanStack Start, Python, Flask, GLM-4 Vision, PostgreSQL, Recharts