/
Mateus Ataide
🇧🇷 recife, pernambuco, brasil
ai product builder @ extreme group (volund) | information systems @ ufpe (cin)
software engineer and ai product builder with 3+ years of experience delivering high-impact systems. currently building ai-native agentic pipelines at extreme group (volund), turning months-long projects into weeks. previously led a fintech squad at amigo tech, sustaining systems that process r$60m+/month. focused on golang, node.js/typescript, python, and applied ai (llms, rag and ai agents in production).
studying information systems at ufpe, one of brazil's top cs schools.
really interested in meeting new people and learning about their experiences. reach out if you want to chat!
key metrics
- R$60M+/month processed
- 60% reduction in human interaction via AI agents
- 30% projected revenue increase via ecommerce integration API design and launch
- ~85x query performance improvement (13s → 150ms)
- MED (Pix dispute) system delivered in 1 week
- 4s processing with 20,000 records (Go pipeline migration)
- extreme group (volund) — ai product builderjun 2026 – present
- — transform manual projects into ai-native workflows through specialized agents, eliminating human bottlenecks in repetitive development and operational steps.
- — contributed to building volund os, an agent creation platform for cross-functional use (commercial, product, c-levels), integrating claude code and cursor sdk to unify models and integrations across the organization.
- — integrated a legacy prodesp project from prototype to gcp deploy, reducing human interaction in the development cycle by 60% via ai agent automation.
- amigo tech — software engineer & team leadnov 2025 – may 2026
- — Led a fintech squad, driving technical decisions, solution planning and product alignment, always focused on ownership and delivering real business impact.
- — architected e-commerce integration api (nestjs/typescript + redis) with projected +30% revenue impact
- — architected the med (pix dispute) system from scratch in nodejs and angular, delivered in 1 week with automated tests, webhooks and full flow, meeting a critical bacen requirement — impacts 75% of monthly transactions.
- — led migration of critical pipeline from NodeJS to Golang, reducing processing to ~4s with 20,000 records per execution.
- — operated cloud infrastructure on aws (ecs, s3, sqs, sns, lambda, cloudwatch) with authentication, authorization and access control across payment apis.
- — used claude code daily as an agentic pipeline for planning, implementation, review and debugging.
- amigo tech — software engineersep 2024 – nov 2025
- — built and maintained backend apis and systems in nestjs/typescript and go for a platform processing r$60m+/month, with authentication, authorization and api security best practices.
- — built ai customer support bot using python + rag + openai api — reduced response time by 40% in production.
- — reduced critical postgresql query latency from 13s to 150ms (~85x) via data analysis and profiling-based optimization.
- — implemented observability with grafana and datadog — metrics, logs and alerts — ensuring 99% slo and supporting production experiment analysis.
- brightlift — founding engineer2023 – 2026
- — built and maintained ai agents integrated with llm apis (openai, gemini), implementing rag, mcp, function calling and streaming with focus on maximizing value per token in automation workflows.
- — built llm evaluation and benchmarking infrastructure for model comparison without altering the application layer — enabling data-driven decisions.
- — developed backend with fastapi (python), scraping + nlp pipeline for large-scale data analysis in mongodb, containerized with docker + ci/cd on aws.