What tech stack does Revolut use?

Revolut's backend is JVM-based, built primarily in Java (17/21) and Kotlin, structured as a large fleet of independent microservices. These run on Google Cloud Platform — confirmed by a public Google Cloud case study — orchestrated with Kubernetes and packaged with Docker.

A notable architectural choice: rather than adopting Kafka, Revolut built an in-house EventStore — a stateless, Kotlin/Ktor application backed by PostgreSQL (with master/replica replication) that processes more than 37 billion streamed events a month, citing easier maintenance and customization. Persistence leans on PostgreSQL (queried via jOOQ, migrated via Flyway) with Redis for caching, and observability runs on Grafana, Prometheus and New Relic.

The web frontend is built with React and TypeScript, also deployed in a microservices style. The consumer experience centres on native iOS and Android mobile apps, which are the primary surface for nearly all of Revolut's 68M-plus customers.

On the data layer, PostgreSQL is the system of record, with Redis for caching and the bespoke Postgres-backed EventStore handling inter-service messaging. Public signals on specific GTM/CRM/martech tooling are limited, so those are intentionally omitted here rather than guessed — Revolut is a build-leaning shop and discloses comparatively little about commercial SaaS it buys.

Revolut is a strong-engineering, build-versus-buy-leaning organisation — its decision to build an in-house EventStore instead of adopting Kafka is a clear tell that it will build core infrastructure rather than buy it. Pitches that try to replace primitives it already owns (messaging, core data) will face an uphill battle.

The higher-probability wins are around the edges of a Java/Kotlin + GCP + Kubernetes shop: developer productivity, observability that complements Grafana/Prometheus, security and compliance tooling, cloud cost optimization, and data/ML platforms that complement GCP. Lead with deep technical credibility and proof of scale — this is a buyer that will scrutinise architecture and prefers tools that slot into a cloud-native, microservices world.