Building scalable, fault-tolerant systems using functional programming and high-performance runtimes — with production experience in card authorization, regulatory reporting, and fraud detection at scale.
Specialized in Go, Clojure, and Elixir, with hands-on depth in distributed architectures, concurrent systems, and database internals.
Backend engineer focused on systems where correctness isn't optional — payment authorization, ledgers, and fraud detection, where a bug isn't just a bug, it's a financial or regulatory incident.
Professionally, I've worked on ISO 8583 message processing for card authorization, regulatory reporting (CADOC, DIMP) for financial compliance, and fraud detection at Nubank's scale. Outside client work, I build the infrastructure layer most engineers treat as a black box — most recently, a sharded key-value database from scratch to understand how consistency guarantees actually work under concurrent writes.
I gravitate toward functional programming for the reasoning guarantees it gives you around state and concurrency, and I care more about systems staying simple, correct, and reliable under load than about using the newest tool.
Core interests
- Distributed systems & database internals
- Payment systems & financial infrastructure
- Functional programming
- High-performance, concurrent APIs
- Event-driven architecture
Currently exploring
- Streaming systems (Kafka / Flink)
- Service mesh & observability
- Cloud-native infrastructure
📍 Brazil — open to remote opportunities worldwide
Languages
Go Clojure Elixir Python Kotlin TypeScript
Backend & Frameworks
Phoenix Pedestal Django Node.js
Databases
PostgreSQL Datomic Redis MongoDB
Infrastructure
Docker GitHub Actions CI/CD REST APIs Microservices
Domain
ISO 8583 Payment Authorization Regulatory Reporting Fraud Detection
A sharded key-value database built from scratch in Elixir to explore database internals, concurrency models, and distributed storage patterns firsthand.
Core features
- Consistent hashing with parallel shards
- ACID transactions
- MVCC (Multi-Version Concurrency Control)
- Write-Ahead Logging (WAL)
- Bloom filters for read optimization
- In-memory cache layer
- HTTP API interface
Stack: Elixir · Plug/Bandit · Docker Repo: github.com/josuesantos1/cube
- GitHub: github.com/josuesantos1
- LinkedIn: linkedin.com/in/jossdev
- Blog: joss.fly.dev
"Programs must be written for people to read, and only incidentally for machines to execute." — Harold Abelson