Backend Architecture & Software Development

Microservices, modular monoliths, Kafka and Apache Beam — from architecture to handover.

We design and build backend systems: microservice and modular monolith architectures, event-driven communication with Kafka, stream and batch processing with Apache Beam, and the APIs, interfaces and operational layer around them. In production this has included a platform at 340 million data points per day, built on Java/Spring Boot and Go services with Kafka, Beam, Redis and InfluxDB.

We take an area end to end — architecture through implementation to production — and stay through the phase where the design meets real load. Tests, pipelines, documentation and handover are part of the engagement, since the measure of the work is whether your team can carry it afterwards.

Engagement model. Longer engagements, typically 3–12 months, part- or full-time. Remote or on site in the Munich area. German and English. We work inside your team and your repository, with responsibility for an area end to end.

Our Focus Areas in Backend Architecture

  • Architecture
    Service boundaries, deployability and data ownership: microservices where the boundaries are real and teams need to ship independently, a modular monolith where the coordination cost would exceed the benefit. A decision made per system, not a default.
  • Event-driven systems with Kafka
    Asynchronous messaging and decoupling between services, topic and event design that tolerates schema change, and explicit handling of ordering, idempotency, replay and unprocessable messages.
  • Stream and batch processing
    Apache Beam pipelines over stream and batch on the same data, including batching for efficient ingest at volume. Temporal for long-running, stateful workflows spanning systems.
  • APIs and contracts
    REST and GraphQL, documented with Swagger: field ownership, error semantics and versioning defined before integration rather than after.
  • Services, persistence and interfaces
    Go, Java/Spring Boot, Python/FastAPI, Node.js/NestJS, Java/Quarkus. PostgreSQL and MongoDB, Redis for caching, InfluxDB and TimescaleDB for time series. On top: dashboards, internal tools and portals in Vue 3/Nuxt or React, and Flutter for mobile — including the WebView approach that lets web and mobile share one implementation.
  • Operations and modernisation
    CI/CD with GitHub Actions and GitLab CI, Docker and Kubernetes on AWS and GCP, observability with Prometheus, Grafana and Datadog, automated testing with JUnit and Jest. Technical debt reduced alongside feature work, and framework migrations in live systems — including a full Vue 2 to Vue 3 migration in production with Pinia and a TypeScript component library in Storybook.

In production

  • B2B2C energy platform
    Microservice architecture in Java/Spring Boot and Go with REST and GraphQL APIs, Kafka for asynchronous messaging, Apache Beam for data pipelines, Redis for caching, InfluxDB for time series. 340 million data points per day; growth from 700 to 20,000 users; technical debt reduced by around 30%. Deployment on AWS with CI/CD, Docker and Kubernetes, observability in Prometheus, Grafana and Datadog.
  • Internal workflow management system
    Modular backend services in Python/FastAPI and Java/Quarkus in a service-oriented architecture, Kafka decoupling the components, PostgreSQL for relational data, REST APIs, and a React frontend for complex workflow screens and dashboards.
  • MedCircle (medcircle.eu) — our own platform
    Connects medical students in Germany with clinics and hospitals. Architecture, NestJS backend with REST APIs and JWT authentication, Vue 3/Nuxt frontend in TypeScript with Pinia, MongoDB. Built, launched and operated by us; in production with 50+ doctors and 100+ students.

Technologies: Go · Java/Spring Boot · Python/FastAPI · Node.js/NestJS · Java/Quarkus · Kafka · Apache Beam · Temporal · REST · GraphQL · Swagger · PostgreSQL · MongoDB · Redis · InfluxDB · TimescaleDB · Vue 3 · Nuxt · TypeScript · React · Flutter · Docker · Kubernetes · AWS · GCP · GitHub Actions · GitLab CI · Prometheus · Grafana · Datadog · JUnit · Jest

Describe the system and the constraint behind it.

You get a direct read on scope and approach.