André Vinícius

AI Engineer & Software Engineer building Multi-Agent Systems, Machine Learning models, and scalable distributed backends.

About Me

I am a Software Engineering Intern at the Brazilian Federal Court of Accounts (TCU), an Undergraduate AI & Recommender Systems Researcher at UEFS, and a Google Student Ambassador 2026. Specializing in Python, TypeScript/Next.js, Go, and Elixir, I build production-grade Multi-Agent LLM architectures (Claude Agent SDK, MCP, deterministic guardrails, τ-bench evaluation), approximate nearest-neighbor recommendation engines (LSH Forest & MinHash), and fault-tolerant distributed backends.

André Vinícius

Featured Engineering Projects

See all projects →

Comanda — Multi-Agent Delivery Customer Support

Claude Agent SDK · Model Context Protocol (MCP) · FastAPI · Next.js · SQLite

Multi-agent orchestration (Reception on Claude Haiku + Order & Refund Specialists on Sonnet) with per-agent MCP servers, deterministic refund calculation in code, 4 tool guardrails, custom Portuguese BM25 + Snowball stemmer from scratch, and a 30-scenario τ-bench evaluation suite (8 adversarial, LLM-as-a-Judge, pass³) achieving 0 policy violations at $0.026/session.

LSH Forest & MinHash Clinical Recommender

Python · Datasketch · LSH Forest (l=200, k=20) · Serverless LRU Cache · MongoDB

Production User-Based Collaborative Filtering engine built at UEFS. Compresses patient histories into 200-permutation MinHash signatures indexed in self-tuning LSH Forest prefix trees, with a Memory-First LRU cache (~200ms warm start), batch inference (300 patients/call), deterministic clinical gender filtering, and 4-level graceful degradation.

Flow — High-Throughput Telemetry Pipeline

Elixir/OTP · Phoenix · Broadway · Oban · PostgreSQL

Asynchronous event ingestion and processing pipeline using Elixir/OTP and Broadway with configurable batching and backpressure, persistent background jobs via Oban, real-time Phoenix PubSub broadcasting, and Telemetry Metrics instrumentation.

Professional Experience

Logo da TCU

Tribunal de Contas da União (TCU)

June 2026 – Present

Software Engineering Intern

- Develop and maintain enterprise backend features and business rules utilizing Apex, PL/SQL, and JavaScript to support federal auditing and oversight platforms.
- Optimize complex relational SQL queries and persistence routines with a strong focus on performance, reliability, and data integrity.
- Investigate production issues and implement fixes across mission-critical corporate applications.

Applied Research & Engineering

Logo da UEFS

Universidade Estadual de Feira de Santana (UEFS)

Undergraduate Researcher — Software Engineering & Recommender Systems

September 2025 – Present
- Engineered LogicFarma, a pharmaceutical management SaaS built with Next.js, Node.js, and MongoDB following Clean Architecture principles (4,700+ operational records and 1,060+ medications).
- Designed and deployed in production a Python recommendation engine using LSH Forest and MinHash (l = 200 permutations, k = 20 neighbors) with a Memory-First Serverless LRU cache (~200ms warm-start latency).
- Implemented batch inference (up to 300 patients per HTTP request), deterministic clinical gender filtering, and 4-level graceful degradation (LSH → Partial LSH → Segmented Global Fallback → Timeout Fail-safe).

Leadership & Extracurricular Activities

Logo do Google

Google

August 2026 – Present

Google Student Ambassador 2026

- Lead campus initiatives to upskill university students in practical applications of Google AI technologies and Gemini workflows.
- Conduct technical knowledge-sharing sessions on modern development tools and generative AI integration.

Awards & Recognitions

Logo da Claro

Instituto Claro / LSITEC

National Finalist — 13th Edition of Campus Mobile

Dec 2024 – Feb 2025
- Selected as a National Finalist in the Health category for the 13th edition of Campus Mobile, pitching an original tech MVP to healthcare executives and engineers.
- Developed a high-engagement Minimum Viable Product (MVP) and refined product architecture through intensive mentoring at USP/LSITEC.

Courses and Certifications

Logo da CesarSchool

CESAR School

Logo do Instituto Claro

Instituto Claro / LSITEC

Technical Articles

Recent Publications

Deep dives into Multi-Agent LLM orchestration, Model Context Protocol (MCP), Recommender Systems in production, and Neural Networks from scratch.

AI Engineering & Multi-Agent Systems

Como construir agentes de atendimento que não alucinam dinheiro: regras em código, guardrails MCP e avaliação τ-bench em português

Arquitetura do Comanda: orquestração multi-agente sobre Claude Agent SDK e MCP com cálculo determinístico de reembolsos em código, guardrails antes das ferramentas e avaliação automatizada estilo τ-bench com 30 cenários.

AI AgentsMCPPython
Machine Learning & Recommender Systems

Sistemas de Recomendação em Produção com LSH Forest, MinHash e Cache LRU Serverless (latência de 200ms)

Como projetei e coloquei em produção na UEFS um motor de recomendação por Filtragem Colaborativa com LSH Forest (l=200), MinHash, Cache LRU em memória e degradação graciosa em 4 níveis.

Machine LearningLSH ForestMinHash
Machine Learning & Neural Networks

Implementando Redes Neurais e Inteligência Computacional do zero em Python (Perceptron, Adaline, MLP, Kohonen, Fuzzy e Algoritmos Genéticos)

Como implementar a matemática fundamental de Redes Neurais Artificiais (Perceptron, Adaline LMS, MLP Backpropagation, Mapas de Kohonen) e Inteligência Computacional sem usar frameworks caixa-preta.

Neural NetworksMachine LearningPython

Core Tech Stack

PythonTypeScriptClaude Agent SDK & MCPLLM Evaluation (τ-bench)LSH Forest & MinHashBM25FastAPINext.jsNestJSGoElixir / OTPPostgreSQLMongoDBSQLiteDocker
Get in Touch

Let's build something together

Feel free to reach out if you're looking for an AI / Software Engineer, have a question, or want to connect.

andrevinciusdiz@gmail.com