Public case

LogTech and last-mile

Tracejo

Last-mile operating system with AI routing, driver app and a real-time control tower

Next.js 15TypeScript.NET 8PostgreSQLRedisMapboxKafka
Tracejo
Role
Tech Lead and Software Architect
Stack
Next.js 15 • TypeScript • .NET 8 • PostgreSQL • Redis • Mapbox • Kafka

Summary

SaaS platform for last-mile and e-commerce operations with modules for AI routing, driver mobility, control tower, and multi-contractor billing.

Problem

Delivery operations in Brazil still rely on shadow spreadsheets, location apps disconnected from billing, and manual routing. Tracejo was built to unify this into a single platform with real SLA.

Scope

I led the full product architecture: VRP routing engine, Android/iOS driver app with offline mode, live control tower, multi-contractor billing module, and a public SDK for integrators.

Impact

1,842 fleets in production, an average 38% reduction in cost-per-delivery after 90 days, and a 97.8% SLA sustained across operations of 50 to 800 active drivers.

Architectural decisions

Modular .NET 8 backend with Kafka queues for route events, region-partitioned PostgreSQL, Redis for driver-in-route state, and Mapbox for tiles and ETAs. Next.js 15 + Server Actions for the tower UI. Mobile app in React Native with native modules for location and proof-of-delivery scanning.

Tags

LogTechLast-mileRoteirização IAB2B
Permanent link

Gallery

Tracejo hero featuring the live operations tower, routes and real-time map.

Tracejo hero featuring the live operations tower, routes and real-time map.

Tracejo mobile experience with value proposition and trial CTA.

Tracejo mobile experience with value proposition and trial CTA.

Full Tracejo page — metrics, modules, API, testimonials and pricing.

Full Tracejo page — metrics, modules, API, testimonials and pricing.

Next step

If the challenge demands technical clarity and real delivery capability, I step in to unlock it.

Discuss a project