Service

Backend Engineering

Backend engineering for transactional and data-intensive systems — domain modelling, API design, integration work, and the reliability engineering that keeps it upright.

Core stack
Java · Spring Boot · Kotlin · Node.js · PostgreSQL

Correctness first, cleverness rarely

Backend defects are expensive because they are quiet. A double-charged payment, a lost message, a race in a state machine — these surface as support tickets weeks later, after the data has already spread.

We design for the failure cases explicitly: idempotency keys, transactional boundaries, retry semantics, dead-letter handling, and observability that makes the state of a request reconstructable after the fact.

Typical work

  • Extracting a coherent service out of a tangled monolith
  • Designing an event model that does not require distributed transactions
  • Making a batch process incremental so it stops missing its window
  • Building integration layers against legacy or badly documented third parties

Case studies

Where this work shows up

Problem, constraint, decision, consequence. No vanity metrics, no logos we are not allowed to name.

All case studies

Sample contentSample engagement. Illustrative scenario used while the site is in build — not a named client project.

fintech2025
  • Java
  • Spring Boot
  • Kotlin

01European payment service provider (sample scenario)

Payment Platform Modernization

A ten-year-old payment monolith moved from quarterly release windows to weekly deployments, without a rewrite and without a settlement incident.

  • Release cadence moved from quarterly windows to weekly deployments
  • Reconciliation lag reduced from overnight to near real time
  • Java
  • Spring Boot
  • Kotlin
  • PostgreSQL
  • Kafka
data and analytics2025
  • Java
  • Kafka
  • PostgreSQL

02Analytics provider in the retail sector (sample scenario)

High-Load Data Processing System

A nightly batch pipeline that had outgrown its window, re-architected into an incremental streaming system with correctness checks that alert before customers notice.

  • Processing moved from a single nightly window to continuous incremental updates
  • Single-day reprocessing without re-running the full pipeline
  • Java
  • Kafka
  • PostgreSQL
  • Redis
  • Docker

Start a conversation

Need this kind of work?Start here.

Tell us what you are trying to build, fix or decide. If we are not the right team for it, we will say so and point you somewhere better.

Typical first step: a 30-minute call, then a short written assessment.