Remote: Yes. PST timezone preferred.
Willing to relocate: No, but open to occasional travel to an office within reason.
Résumé/CV: https://drive.google.com/file/d/1gGjvZMnUQ3EVj3GeSlWt8TIpg6w...
Email: andrew@deandrade.com.br
I’m a backend-focused engineer with deep experience building highly distributed, high-throughput stateful and stateless systems - millions of requests per second across thousands of machines in multiple datacenters. My strongest skillsets are around reliability, performance, and testing at scale: I’m an expert in CI/CD, software testing, and the guardrails that let teams move fast without breaking production. I'm a world-class level expert in end-to-end testing in massive distributed systems.
Technically, I’ve worked most with Golang and Javascript/Typescript, but also with Python, Java, and Lua professionally, and Lisp/Haskell for fun. I’d prefer to work in Go, but I’m open to any language I’ve touched or learning something new. If a technology has documentation, I’ll pick it up quickly; if not, I’ll dive into the source. I’m primarily a backend engineer, but I’m not a stranger to frontend work.
I spent just shy of 10 years at Uber (2014–2024), where I was a staff engineer working on some of Uber’s largest, most complex stateful services. Before that, I was at a startup from seed through Series B as engineer #1. I’ve operated comfortably in environments from early startup through large-scale enterprise.
What I’m looking for: I want to work at a company that is seriously embracing AI-assisted software engineering. From personal projects and experimentation, I’ve seen how AI tools feel like a team of fast, naive junior engineers. They need strong production engineering practices - reproducible build systems, tests, linting, typed languages, static analysis, structured logging, metrics, integration and end-to-end tests - to keep velocity from becoming fragility. My goal is to bring that discipline to teams so AI-assisted development produces robust, production-ready systems, not just rapid prototypes.
Here's a link to my recent portfolio experiment in building a very simple CRUD application, but with all the features I mentioned in the previous paragraph that enable velocity, linting, etc: https://github.com/andrewdeandrade/cc-explore
For the recruiters, you can take a look at the progress/ directory for AI summaries of 4-days worth of work in this repo.
I'm a long-time HN member with almost 9000 karma/points, but posting on a new account to maintain some degree of anonymity for my primary account.