Sergey Vareyko / Logistics Software

Logistics software development: transportation management and freight platforms

This page is for shippers, freight brokers, carriers and logistics providers whose work is spread across phone calls, e-mails and disconnected tools: orders, carriers, shipment status, documents and billing need to live in one system that every party can work in. I'm Sergey Vareyko, a software architect with 15+ years in Java and Spring. My logistics software development work covers a cloud transportation management system (TMS) that connects every participant in a shipment, and a freight management platform that picks carriers from the data it has accumulated. Below: what I build, how I work and the projects behind it.

What I build

Everything below comes from the platforms described further down: Turvo, a collaborative TMS, and Gulf Relay, a freight management platform.

  • One workspace for every party. Shippers create transportation requests with cargo specifications, delivery requirements and billing details; freight brokers coordinate carriers; carriers manage execution and post delivery updates. Every authorized participant follows the same shipment and exchanges information and documents on it.
  • Shipment management end to end. Transportation orders, requirements, routes, schedules and billing in one place, with workflow automation for the repetitive steps.
  • Real-time tracking. Shipment locations, delivery progress and status updates, visible to everyone who is authorized to follow the shipment.
  • Carrier matching from your own data. Suitable carriers are identified automatically from historical shipment data, carrier performance and each customer's selection criteria, gathered in a centralized data lake, instead of relying only on manual selection.
  • Rules per customer. Flexible business rules and personalized matching criteria, so that different customers set their own priorities and transportation requirements in one platform.
  • Integrations. External logistics services, warehouse and enterprise systems connected through REST APIs and third-party integrations, so that operational data is consolidated in one place.
  • Volume without waste. Search storage that takes less disk space for large volumes of logistics data, CPU-heavy operations on large datasets optimized, and cloud resources reviewed to cut AWS costs.
  • Analytics and reporting. Operational dashboards and business insights for logistics planning and decisions.

Underneath it is Java and Spring. On Turvo I worked with Spring Boot, MySQL, MongoDB, Redis, Elasticsearch, RabbitMQ and GraphQL on AWS; on Gulf Relay with Spring microservices, Microsoft SQL Server, MongoDB and Elasticsearch on Microsoft Azure, with Azure Queue and Azure Load Balancer. The same shape — roles, rules and integrations around one shared record — is behind the custom business systems I build, and the monitoring and cloud-cost side is on the internal tools page.

How I work

I start from how freight moves today: who creates orders and in which systems, how carriers are chosen and on what criteria, which data already exists about past shipments and carrier performance, and who needs to see what. From that I define the structure of the system, the stack, the integrations and how it will scale with volume. That is what I do as a solution architect on new client engagements; on Gulf Relay I led the architectural and engineering work, coordinating implementation decisions and guiding the development of the backend.

Then a working version comes early, and the platform grows through iterations on real orders. A logistics system also needs work once data piles up: on Turvo I optimized how Elasticsearch stored logistics data, reworked CPU-intensive operations on large datasets and reduced AWS expenses by optimizing resource utilization. Whether the platform is new or already running, you receive a system with its integrations connected and backend services built to grow with your volumes.

Projects in this domain

Turvo — Software Architect and Lead Software Engineer, 2020–2021. A cloud-based transportation management system that connects shippers, carriers, freight brokers and logistics providers in one collaboration platform: shipment planning, real-time tracking, workflow automation, document exchange, integrations with logistics, warehouse and enterprise systems, and operational dashboards. I optimized Elasticsearch data storage, which significantly reduced the disk space needed for large volumes of logistics data, designed and developed a new microservice with a combined storage approach, optimized CPU-intensive operations on large datasets, and reduced AWS expenses by optimizing resource utilization.

Gulf Relay — Software Architect and Team Lead, 2022–2023. An enterprise logistics and transportation management platform that automates freight operations and order processing and chooses carriers from data: it identifies suitable carriers from historical shipment data in a centralized data lake, carrier performance and customer-specific criteria, under business rules each customer can define. I designed and implemented its highly scalable microservice-based backend services, the foundation for processing complex transportation workflows and growing data volumes, and led the architectural and engineering activities.

Altir Foundation — 2021–2022. Reusable components and services of a software product studio, one of whose application areas is supply chain and logistics automation: shippers, freight brokers and carriers coordinating transportation in real time and tracking shipments through web and mobile interfaces. My part was the shared foundation: an Events microservice with a Complex Event Processing engine that correlates and reacts to event streams, an Elasticsearch integration library with automated migration management, and the design of centralized authentication.

Frequently asked questions about logistics software development

Can the system choose carriers for us?

It can do most of the selection. On Gulf Relay suitable carriers are identified automatically from historical shipment data, carrier performance and each customer's own criteria, instead of relying only on manual selection, and every customer's priorities are expressed as business rules and matching criteria. How much of the final decision stays with a person is a business choice made at the start.

Should we build a transportation management system or adapt a ready-made one?

A ready-made TMS fits when your process fits it. A custom build pays off when the way you choose carriers, the parties you work with or the data you have accumulated is your advantage, or when the system has to talk to services a product does not know. A custom platform can start with the core order and shipment flow, with matching and analytics added on top.

Can it connect to our warehouse and enterprise systems?

If they have an API or an exchange format, yes. Turvo integrates with external logistics, warehouse and enterprise systems, and Gulf Relay with external logistics services and enterprise systems. Each integration is a separate piece of work, so the list of systems you already run is the most useful thing to bring to the first conversation.

How long does it take?

It depends on scope: the number of parties and roles, the integrations, how much historical data has to be loaded, whether carriers need a mobile interface, and the volumes the system must handle. I estimate once the requirements are on the table.

Where does it run, and will it cope with our volumes?

In the cloud of your choice: Turvo ran on AWS and Gulf Relay on Microsoft Azure. Volume is part of the design from the start: Gulf Relay's backend was built as scalable microservices for growing data volumes, and on Turvo I reduced the storage and CPU cost of large logistics datasets. Hosting, the repository and support are agreed at the start.

Need an engineer with experience in logistics software development? Write to me — a few lines about the task are enough — or look through my CV.