RS recep.dev

Available for backend work

Backend engineer building
systems that stay up.

I design and ship distributed services, APIs and data pipelines where correctness and latency actually matter. Ten years of turning ambiguous requirements into boring, reliable infrastructure.

View work Get in touch
10+
Years shipping
99.98%
Uptime owned
40M
Requests / day
12
Services in prod

01 — Selected work

Things I built and still maintain

Platform

Event ingestion pipeline

Kafka-backed ingestion handling 40M events per day with exactly-once semantics, schema registry enforcement and replayable dead-letter queues.

  • Go
  • Kafka
  • Postgres
  • Kubernetes
Security

Multi-tenant auth service

OAuth2 and OIDC provider with per-tenant key rotation, fine-grained scopes and sub-10ms token verification at the edge.

  • Rust
  • Redis
  • OIDC
  • Terraform
Observability

Latency budget toolkit

Internal library that traces every request across services and fails the build when a p99 budget regresses. Cut incident triage from hours to minutes.

  • Python
  • OpenTelemetry
  • Prometheus
  • Grafana
Data

Zero-downtime migration runner

Online schema migration tool for multi-terabyte Postgres clusters, with automatic backfill throttling and instant rollback.

  • Go
  • Postgres
  • gRPC
  • AWS

02 — Stack

What I reach for

Languages

  • Go
  • Python
  • Rust
  • TypeScript
  • SQL

Infrastructure

  • Kubernetes
  • Terraform
  • AWS
  • Docker
  • GitHub Actions

Data

  • PostgreSQL
  • Kafka
  • Redis
  • ClickHouse
  • Elasticsearch

Practice

  • Distributed systems
  • API design
  • Observability
  • Incident response
  • Mentoring

03 — About

Boring infrastructure, on purpose

I'm a backend engineer focused on the unglamorous parts of software: the retry logic, the migration that doesn't page anyone at 3am, the API contract that survives its third consumer.

Most of my work sits at the intersection of distributed systems and developer experience. I like small interfaces, explicit failure modes and systems you can reason about on a whiteboard.

Outside of work I write about reliability engineering and contribute to open-source tooling around observability.

04 — Contact

Let's build something that holds up.

Open to backend and platform engineering work, consulting, or just a good conversation about distributed systems.

hello@example.com GitHub LinkedIn