About Portfolio Cases Services Blog Contact 🎙 Talk to AI
EN DE RU
🎙 Talk to AI
Case study · Production SaaS

Digital Manager

“A phone-first operations platform (PWA) for multi-location retail: photo-verified task checklists, role-based team chat, shift alerts and a network “discipline index” — replacing spreadsheets, paper logs and scattered messengers.”

Architected & built by Denis Shokhirev · Agentic AI Systems Architect, Freiburg
● Production HoReCa · Retail
≤500
Locations / instance
20
Photos per task
4
Chat threads
3
Roles
The problem

Running shift staff across a distributed retail network by hand — spreadsheets, paper logs, and chats scattered across several messengers. Routines get forgotten, there's no photo proof that work was done, and the owner has no visibility into a location when they're not standing in it.

What I built

A phone-first PWA on React + Supabase with database-level location isolation (RLS). Photo-verified task checklists (up to 20 photos per task, auto-compressed, with an offline queue so nothing is lost on a bad connection), a four-stream team chat (team / shift-handover / announcements / system), roles for owner, manager and staff, an in-app knowledge base, and web push that works without any messenger or app store. A network “discipline index” rolls completion up across locations. The architecture scales to ~500 locations per instance.

React PWASupabasePostgreSQL + RLSRealtimeWeb Push (VAPID)Object storage

A camera / computer-vision module and an AI assistant for the manager are on the Phase-2 roadmap.

Operations, digitized

Run every location from your phone.

If you manage shift staff across multiple locations and live in spreadsheets and group chats — there's a better way.

Start a project → More cases