Unlocking 3x scale capacity with a custom TMS built in 6 months

Overview

Industry

Location

Transportation and logistics

Poland

Technology Used

Angular

Java

Kotlin

Project overview

The client’s legacy off-the-shelf system became a growth blocker, incapable of scaling beyond 40,000 shipments per day with operations fragmented across several disconnected tools. We replaced the failing software with a custom Transport Management System built in just six months – a fraction of the 18–36-month industry-standard timeline. Using an AI-first methodology and a small autonomous team, we delivered an integrated command center and a native mobile app for 600+ couriers covering the full last-mile stack: automated geocoding with dispatcher correction, multi-stop route optimization, digital proof of delivery, hub management, real-time fleet tracking, failed delivery handling and 3-click delivery confirmation – the same operational core that drives CEP networks at scale.

Client background
One of Poland’s largest direct-to-consumer operators, running a nationwide last-mile cold chain network of 600+ couriers. The company manages 60,000 daily shipments with fixed overnight delivery windows and high-volume logistics requirements.
Goals
The goal of the project was to replace the client’s legacy system that hindered growth with a high-performance custom TMS. We aimed to unify operations, replacing a patchwork of several disconnected tools with a single, coherent source of truth to reduce manual tasks and provide management with real-time visibility across the daily operational workflow. By leveraging an AI-first development approach, while combining deep CEP domain expertise with leading AI models, the focus was on delivering a feature-rich, future-proof last-mile delivery software for 600+ couriers within a strict six-month timeline and a fixed budget.

Results

50% faster geocoding and routing
The system cut processing time in half, allowing planners to refine routes and improve delivery quality within tight nightly windows.
70% reduction in manual steps
By unifying fragmented workflows into one, coherent source of truth, the vast majority of re-keying and informal workarounds was eliminated.
3-day feature deployment
The time required to launch new business features dropped from months to just three days, allowing the company to adapt at market speed.
100% system uptime
The solution has ensured zero operational disruptions since launch, providing a fundamental shift in reliability for an operation with no margin for error.
50% faster courier onboarding
The mobile app’s user-centric design — built with couriers in the field, not just in the office — enables a fleet of 600+ drivers to get operational in half the time. In high-turnover last-mile environments, this directly translates to lower training costs and fewer errors during the critical first weeks on route.
Built-in 3x volume headroom
The new architecture was stress-tested and built to accommodate three times the current shipment volume, ensuring it can support aggressive growth without future re-platforming.
100% data centralization for AI
Previously fragmented operational data is now centralized in a clean, consistent format designed to support future machine learning and LLM-powered features.
6-month delivery timeline
The entire custom TMS was delivered in six months – a fraction of the 18–36-month industry standard – while adhering strictly to the fixed budget.