Panels
Commissioned by the IEA study center (Bureau d'Études), Panels was created to solve an unoptimized research workflow that multiple companies failed to deliver. Developed by myself and the Bayna team, we were the first to successfully engineer a working solution. Panels is a local, AI-assisted platform that processes hundreds of receipt and product images—each containing dozens of items—on constrained computational hardware, accelerating researcher workflows while guaranteeing 100% local data privacy.
About This Project
The IEA study center struggled with an unoptimized data workflow that multiple previous software providers failed to solve, requiring secure local processing of complex multi-product images under low computational power.
Collaborated with the Bayna team to engineer a high-performance local AI pipeline using Tauri and Rust, becoming the first to successfully digitize and automate the IEA research workflow.
A breakthrough data processing platform for market research that processes multi-item receipts locally without cloud reliance.
Full-stack Developer
2025
Private
Client
Technology Stack
Project Story
The IEA study center struggled with an unoptimized data workflow that multiple previous software providers failed to solve, requiring secure local processing of complex multi-product images under low computational power.
Collaborated with the Bayna team to engineer a high-performance local AI pipeline using Tauri and Rust, becoming the first to successfully digitize and automate the IEA research workflow.
A breakthrough data processing platform for market research that processes multi-item receipts locally without cloud reliance.
Insights & Takeaways
Highlights
- Commissioned by the IEA study center; first team (alongside Bayna) to successfully solve their complex workflow.
- Batch processes hundreds of receipt images with dozens of products per photo locally.
- Engineered for low computational power hardware with 100% offline data security.
Challenges
- Structuring local batch processing queues for multi-item images under strict memory limitations.
- Ensuring high OCR accuracy across varied receipt print qualities and lighting.
Lessons Learned
- Combining local AI inference with efficient desktop frameworks enables enterprise-grade data processing without expensive cloud infrastructure.
Related Work
Case Study
Auto Shutdown
Rust + React desktop app to automate PC shutdowns by timer, schedule or idle state.
Case Study
Azkari
Lightweight desktop app that displays Azkar reminders based on time of day, built with Tauri, SolidJS, TypeScript and Rust.
Case Study
EBEC
A full stack, highly creative web platform built for the ENSIA Business & Entrepreneurship Club, complete with an administration and user management system.
