Route planning and evidence in the messenger

For Tabel, a container cleaning service provider operating across Germany, we developed an integrated platform for route planning, job documentation and billing preparation.

Decentralised cleaning teams document their jobs at the respective sites. The platform brings together route planning, job documentation and billing preparation for decentralised cleaning teams. The cleaners speak many different languages and have no company laptop or desk at the site, while customers require reliable records of cleaned containers and services delivered.

Instead of introducing an additional app, the solution uses a messenger: onboarding is passwordless via the registered mobile number. All interactions in the Telegram bot take place in the language stored in the employee profile. It supports 13 bot languages, with German as the fallback language.

At the site, cleaners scan the relevant container's QR code using a mini app embedded in Telegram. QR check-in records their arrival at the site. Evidence photos must be taken with the live camera; the mini app captures photos with a server-side timestamp and, where location permission has been granted, the device position. Missing or unusual location details are flagged for review. Damage, blocked access or additional services encountered on site are reported directly through a guided dialogue and passed to head office as a ticket.

The dispatch team plans routes and reviews feedback in a shared web interface. Route planning combines geographical proximity, time windows, required qualifications and fixed regular assignments to generate automatic route suggestions.

The dispatch team can understand each route suggestion: transparent scoring explains why a stop is suggested in that sequence. A central image inbox collects all photos received during the day; AI-assisted anomaly detection checks the images in the background and flags unclear or faulty cleaning for manual review.

Integrated operational reporting processes working and job times according to clear rules: statutory breaks and flat-rate travel allowances are taken into account automatically, while absences or forgotten check-outs are flagged and can be corrected in the back office with a reason.

For customers, the system generates weekly service records by customer, site and cost centre, with a detailed comparison of planned and actual completed and aborted cleaning jobs. Data is separated by company; authorised users can access cross-company analyses.

digital cuisine is responsible for the platform's software architecture, concept, full-stack development and ongoing operations for Tabel.

The system is based on Next.js 16, React 19, TypeScript, Bun, PostgreSQL and Prisma with Tailwind CSS and shadcn/ui. The mobile interface uses the Telegram Bot API; CI/CD pipelines and quality assurance run through Azure DevOps.

Period: since 2026 · ongoing operations.