Career

Where I've worked, what I built there, and with what.

  1. Monest · Back-end Developer

    Mar 2026 – Present·7 mos·Curitiba, PR

    • Work on MIA, an AI agent that negotiates debt with millions of people over WhatsApp and voice, handling more than 2 million messages a day.
    • Own agent architecture, evaluation, and observability, including an evaluation pipeline that cut false positives by 5.5×.
    • TypeScript
    • NestJS
    • React
    • Next.js
    • AWS
    • BullMQ
    • Tailwind CSS
    • LangSmith
    • New Relic
    • Terraform
  2. Loomi · Full-stack Developer

    Mar 2025 – Mar 2026·1 yr 1 mo·Recife, PE

    • Designed and shipped Backend-for-Frontend services in Node.js and TypeScript for Vivo, Brazil's largest telecom, supporting high-traffic applications.
    • Architected scalable BFF layers aggregating data from multiple microservices, cutting frontend integration complexity and improving API response time by ~30%.
    • Built AI agents for Brazil's National Council of Justice (CNJ) with LLMs, RAG pipelines, and vector databases, automating legal and administrative workflows and reducing manual review time by 40%+.
    • TypeScript
    • NestJS
    • React
    • AWS
    • Tailwind CSS
    • Langfuse
  3. Sipe · Full-stack Developer

    Oct 2023 – Mar 2025·1 yr 6 mos·Natal, RN

    • Built features for a multi-tenant HR management system in C# and React, enabling centralized operations and halving company onboarding time.
    • Developed a student attendance and class diary system in React, TypeScript, and Node.js for 5 municipal schools, cutting manual administrative work by ~40%.
    • Contributed to a city-council voting platform in full-stack TypeScript and REST APIs, improving transparency and shortening the voting process by 30%.
    • TypeScript
    • React
    • AWS
    • C#
    • React Native
    • Tailwind CSS
  4. IREDE · Back-end Developer

    Jan 2022 – Oct 2023·1 yr 10 mos·Campinas, SP

    • Owned the platform's main microservice, keeping it healthy and reliable.
    • Built a media-processing pipeline on FFmpeg to process and manage room recordings.
    • Used Amazon S3 for file storage and Amazon SQS for messaging between microservices, improving scalability and reliability.
    • Migrated every HTTP route from Gorilla/Mux to the Echo framework, improving performance and responsiveness.
    • Go
    • AWS
    • Echo
    • Vue.js