Software Engineerfull-stackbackend performancedata engineering
Walter
Marroquin
I build and tune software across the stack: product frontends, backend services, data pipelines and the infrastructure under them. Based in San Salvador, working in English with international teams.
- Now
- Software Developer
- at Designli
- Building software since
- 2019
- Local time
- San Salvador, El Salvador
- Next.js
- NestJS
- PostgreSQL
- Docker
- AWS
- TypeScript
- Angular
- .NET Core
- Spring Boot
- SQL Server
- Datadog
- Claude
01Work
Every job on one timeline
Drawn like a request trace: one span per job, roles nested inside. Move across the lanes to scrub through time, pick a span to read what happened there.
Software Developer at Designli
- Build full-stack product features with Next.js on the frontend and NestJS on the backend, backed by PostgreSQL, covering APIs, data models, and complex business logic.
- Develop configurable form workflows with conditional visibility, filtering, and validation rules that support complex product configurations and their dependencies.
- Contribute to deployment and DevOps work with Docker and AWS across the delivery lifecycle.
02Case study
One slow endpoint
At NICE, the API behind the game history took about ten minutes to answer. It was re-running a game simulation it never needed. Computing player results straight from the records already in the database brought it under three seconds.
03Projects
Things I've built and led
This site and its CV system
The site you're on. My CV is written in Markdown in English and Spanish, rendered on the web in nine designs and exported to A4 PDF with headless Chromium. A private editor keeps drafts separate from the published version.
- Next.js
- TypeScript
- Playwright
- Auth.js
Multi-source ETL platform
A data integration platform that extracts, validates and standardizes large datasets from Excel, TXT, JSON, CSV, DBF and SPSS files, with scheduled batch jobs and parallel processing. I was tech lead of a four-person team.
- Java
- Spring Boot
- Spring Batch
- SQL Server
Contact center gamification
Games, real-time challenges, user metrics dashboards and data export for a large-scale contact center platform. Includes the game-history API that went from about ten minutes to under three seconds.
04Lab
What I keep coming back to
Notes and experiments on these will live here.
- Agentic engineering
- How AI agents fit into real delivery work: specifications, implementation, review and validation, and the shared skills that keep a team consistent.
- Backend performance
- Finding the work a system doesn't need to do, and the data access and architecture choices that keep it fast as it grows.
- Data pipelines
- Batch and parallel processing, validation and standardization of messy, multi-format sources.