Sandoval

SANDOVAL, CARLOS DAVID

BACKEND SOFTWARE ENGINEER

Experience
4+ years · backend and full stack
Technologies
Go · Java · Python · JavaScript · PHP
Own projects
Machine learning, graphics and tooling
Education
Bachelor's in Information Systems (in progress)

Backend engineer working on payment systems. On my own I build machine learning, graphics and tooling projects.

About me

I'm a backend software engineer at Mercado Libre, working on Mercado Pago's payment acquiring platform. I specialize in payment systems and transaction control, with a focus on reliability, correctness and scale. I have over 4 years of experience and a previous full stack background in Java/Spring Boot, PHP/Laravel and Node.js.

My day-to-day is Go, MySQL, BigQuery and key-value stores: I design REST APIs and microservices, write automated tests, and use AI tools like Claude and Codex to speed up development. I was also a software quality researcher at UNNE, and I'm studying for a degree in Information Systems. I'm open to backend roles in Argentina and remote worldwide.

Role
Software Engineer at Mercado Libre (Mercado Pago)
Location
Corrientes, Argentina
Languages
Spanish (native) · English (basic)
Carlos David Sandoval

Backend

  • Go
  • Java / Spring Boot
  • JavaScript / Node.js
  • PHP / Laravel
  • REST APIs
  • Microservices

Data

  • MySQL
  • PostgreSQL
  • BigQuery
  • MongoDB
  • Redis

Practices and tools

  • Software testing
  • Git / GitHub
  • Scrum
  • AI tools (Claude, Codex)

Experience

Backend in payments with a previous full stack background. Today at Mercado Libre; before that, research and development at UNNE, drug traceability at Verifarma and full stack web platforms.

  1. March 2022 - Present

    Software Engineer

    Mercado Libre · Mercado Pago (Acquiring)

    • Building business features and rules engines for end-to-end transaction control.
    • Designing, building and maintaining REST APIs and microservices in Go, backed by MySQL, BigQuery and key-value stores.
    • Writing automated tests to keep critical payment flows reliable.
    • Using AI-assisted development tools (Claude, Codex) in daily work.
  2. December 2021 - April 2022

    Software Engineer

    Verifarma

    • Building drug traceability and serialization systems (Track & Trace) used by large multinational pharmaceutical companies such as Bayer and Pfizer.
    • Backend with Java, Spring Boot and MySQL; frontend with HTML, CSS and JavaScript.
    • Designing and integrating REST and SOAP APIs and working with microservices.
  3. March 2020 - November 2024

    Software Quality Researcher and Full Stack Developer

    UNNE · Universidad Nacional del Nordeste

    • Research on backend development best practices (testing, design patterns, clean code) and storytelling techniques applied to software.
    • Development of the group's web platform with Laravel, PostgreSQL and JavaScript: roles and permissions, publications and activity logging.
    • Scrum and iterative delivery; deployed on Heroku.
  4. March 2019 - March 2020

    Full Stack Developer

    Department of Biology · UNNE

    Configurable web system built with Spring Boot and PostgreSQL to publish the department's scientific articles and research: file uploads, roles, publications and email alerts.

  5. June 2018 - June 2019

    Full Stack Developer

    Alfa y Omega

    Vocational orientation test platform with Laravel and MySQL, with payment plans via PayPal and Mercado Pago, dynamic forms (MongoDB) and real-time notifications (Redis).

Education

  1. 2015 - 2027 (expected)

    Bachelor's degree in Information Systems

    Universidad Nacional del Nordeste

  2. 2015 - 2019

    Programming Analyst

    Universidad Nacional del Nordeste

My personal projects

Projects I build outside of work to practice, experiment with new technologies and solve problems I care about.

SR Agro Vision

  • Python
  • PyTorch
  • FastAPI
  • React

A super-resolution application for agricultural satellite imagery. It trains and compares deep learning models (ESPCN, SwinIR and GAN) on 10 m Sentinel-2 images and evaluates the results with agricultural indices such as NDVI. It includes a command-line pipeline and a web application with an API and maps.

It was the most challenging project. The idea came from NVIDIA's DLSS, in its first versions, which used AI to reconstruct low-resolution images in video games. I asked myself how to take that to something that wasn't a game, and I landed on agriculture: letting a farmer enhance free Sentinel-2 images with neural networks instead of paying for commercial high-resolution imagery.

Cornell Box Ray Tracer

  • Python
  • NumPy
  • Path tracing

A photorealistic path tracer built from scratch in Python that simulates the physics of light to render the classic Cornell Box. It supports glass, metal and diffuse materials, depth of field, soft shadows and 3D models with tens of thousands of triangles (such as the Stanford Bunny), accelerated with a BVH structure and parallel rendering.

The idea came from wanting to understand how path tracing works in video games: the math that makes it work and the logic capable of simulating the real environment we live in. And from confirming that everything can be explained with math.

SrvFiles

  • Go
  • React
  • Flutter
  • Tauri

An app to share files between devices over the local network: phone to phone, phone to computer and vice versa, with no internet and no account. It runs on a computer (web or desktop app) or on a phone, and devices connect with a QR code or by discovering the server automatically, to browse, upload, download, preview and transfer files directly to each other with end-to-end encryption. It is two repositories: the web and desktop server, and the Android app.

The idea came from wanting to turn on the app and send files quickly, without resorting to Google Drive, WhatsApp or other temporary channels. Fast, easy and intuitive.

WiXplorer

  • React
  • TypeScript
  • Flutter
  • Go

A file explorer for Android phones that you use from the computer's browser, with no cables or extra software. In WiFi mode, the phone app starts a server with the web interface embedded and you connect with the IP and a PIN; in USB mode, a desktop launcher accesses the phone over MTP. The interface lets you browse, preview and manage the device's files.

The idea came from not finding an easy, direct and intuitive tool that got straight to the point: turn it on, connect to the computer and done. The ones that existed had unfriendly interfaces, were complicated and, often, paid.

Contact

If you want to talk about a role, a project or an idea, write to me. I reply by email.

Location
Corrientes, Argentina