Work

A decade of backend engineering across distributed systems, event-driven architecture, and the kind of scale where process matters as much as code.*

I've been trying to automate myself away for years. Before AI, I'd have spent hours scripting away a 5-minute task in hopes of never having to risk doing something manually twice. Since AI this approach has sped up and grown more professionally viable.

What I do*

I'm an individual contributor working at the intersection of distributed systems and high-throughput data. Day to day that means Kotlin services processing millions of events, Kafka pipelines, and the careful operational work that keeps all of it running cleanly at scale.

I made a deliberate choice to step away from Spring and go vanilla Kotlin. Less magic, more clarity. The same instinct shows up elsewhere: Postgres with JSONB instead of a separate document store, simple concurrency primitives over complex frameworks, bash scripts replaced by Kotlin scripts I can actually read six months later.

At the scale I work at, the biggest wins rarely come from clever code. They come from automation, careful process design, and understanding why a system behaves the way it does. I care a lot about the second-order effects of technical decisions — the ones that only show up after the system has been running for a year.

I've embraced agentic AI workflows as a genuine multiplier for rapid prototyping and exploration, and I think about how to bring that leverage to my team. The language I write in matters less to me than it used to. The architecture and the tradeoffs still matter a lot.

Stack

Primary

  • Kotlin
  • PostgreSQL
  • Kafka
  • RocksDB

Infrastructure

  • Container orchestration
  • Event-driven architecture
  • Distributed systems
  • JVM tuning

Background

  • Java + Spring
  • C#
  • MongoDB
  • SQL Server

How I think*

—

Boring technology is usually the right technology. The interesting problem is the one your system is solving, not the infrastructure underneath it.

—

Reading Designing Data-Intensive Applications reframed how I think about systems more than anything else I've done in my career. I still return to it.

—

A small amount of operational discipline prevents a large amount of production pain. Automation isn't laziness — it's respect for your future self and your team.

—

Deep domain knowledge is a compounding asset. I'd rather understand one domain deeply and learn a new language than understand many languages and no domains.