Skip to content

Miguel de Frias

AI Agent Engineer @ AlphaCorp AI · M.Sc. Computer Science @ University of Brasília (UnB)

📍 Brasília, Brazil

LinkedIn Email

Profile

Software engineer focused on AI systems. I build production-ready AI agents and pipelines using Python, LLMs, and Retrieval-Augmented Generation (RAG), with an emphasis on reliability, retrieval quality, and source-traceable answers. Currently an AI Agent Engineer at AlphaCorp AI and an M.Sc. Computer Science candidate at the University of Brasília (UnB), where I research the foundations and methods of computing.

Professional Experience

AI Agent Engineer — AlphaCorp AI

Aug 2025 – Present · Remote

Build and ship production-ready AI agents and pipelines using Python, LLMs, and RAG. Promoted from Junior AI Agent Engineer (part-time) to full-time AI Agent Engineer. Focus on agent orchestration, retrieval quality, and the infrastructure that makes LLM-driven systems dependable in production.

AI Researcher — CEDIS (Center for Studies, Development & Innovation in Software)

Feb 2025 – Jan 2026 · Brasília · On-site

Led the AI layer of SEI Expert, an AI-powered system that supports the analysis of administrative processes in the Brazilian public sector. Integrated Large Language Models (LLMs), RAG, and the Model Context Protocol (MCP) to deliver context-aware, source-traceable responses grounded in legal documents and process data.

Designed an MCP orchestrator that dynamically builds prompts by aggregating user-profile data, process documents, and relevant legislation — reducing analysts' cognitive load and standardizing decision-making across the organization. Delivered as a full-stack platform: a React/TypeScript frontend, a FastAPI/Python backend, and a PostgreSQL + pgvector database for semantic search.

Stack: Python · FastAPI · React · LangChain · OpenAI API · PostgreSQL · pgvector · Docker · Azure

RAG Researcher — Universidade de Brasília (UnB)

Feb 2025 – Feb 2026 · Brasília

Research into Retrieval-Augmented Generation methods for grounded, reliable LLM outputs.

Teaching Assistant, Software Testing — UnB

Mar 2024 – Feb 2025 · On-site

Supported quality assurance for academic systems through automated testing (JUnit, Pytest, Postman, Jest).

Software Developer — DNIT (Contract)

Mar 2023 – Aug 2023 · Brasília · Hybrid

Developed and maintained an application for Brazil's National Department of Transport Infrastructure (DNIT), delivered in partnership with the University of Brasília. Stack: .NET, C#, React, TypeScript, PostgreSQL.

Teaching Assistant, Engineering Economics — UnB

Sep 2022 – Feb 2023 · On-site

Economic feasibility analysis (NPV, IRR, payback) using Python and financial modeling to support academic decision-making.

Education

  • M.Sc. Computer Science — Universidade de Brasília (UnB) · 2026 – 2028
  • B.Sc. Software Engineering — Universidade de Brasília (UnB) · 2021 – 2025

Certifications

  • Oracle Cloud Infrastructure 2025 AI Foundations Associate (1Z0-1122-25) — Oracle · 2025
  • Oracle Cloud Infrastructure 2025 Certified Foundations Associate (1Z0-1085-25) — Oracle · 2025
  • Additional certifications available on LinkedIn.

Core Competencies

  • AI & Agents: Agentic AI development, AI agents, LLM orchestration, Model Context Protocol (MCP)
  • Retrieval: Retrieval-Augmented Generation (RAG), semantic search, vector databases (pgvector)
  • Backend: Python, FastAPI, LangChain, OpenAI API
  • Frontend: React, TypeScript
  • Cloud & DevOps: Azure, Oracle Cloud Infrastructure (OCI), Docker, PostgreSQL

Selected Projects

  • SEI Expert — AI-powered organizational process analysis for the Brazilian public sector (see Experience).
  • My Market — Software architecture & development project applying requirements engineering, design patterns, and UML. GitHub
  • BRB Mobilidade — Requirements engineering for a public mobility application. GitHub

Languages

Contact

  • 💼 LinkedIn
  • 📧 Professional: miguel.frias@alphacorp.ai
  • 📧 Personal: miguelfrias8@gmail.com
  • 📍 Brasília, Brazil