Back to Home

Projects

Production systems, multi-agent AI architectures, machine learning implementations, and distributed backends.

AI & ML

Comanda — Multi-Agent Delivery Support

Multi-agent orchestration for food-delivery post-order support (Reception on Claude Haiku + Order & Refund Specialists on Sonnet) with isolated MCP servers, deterministic refund rules 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).

PythonClaude Agent SDKMCPFastAPINext.jsSQLite
LSH Forest & MinHash Clinical Recommender

LSH Forest & MinHash Clinical Recommender

User-Based Collaborative Filtering engine in production (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), and 4-level graceful degradation.

PythonLSH ForestMinHashDatasketchNext.jsMongoDB
Backend & Distributed

Flow — Elixir/OTP Telemetry Pipeline

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

ElixirPhoenixBroadwayObanPostgreSQLExUnit
Backend & Distributed

SignalDesk — Multi-Tenant Case Queue API

Multi-tenant backend API in NestJS and TypeScript with JWT authentication, role-based access control (RBAC), transactional concurrency control against race conditions, strict tenant isolation, and Jest/Supertest integration tests.

NestJSTypeScriptPostgreSQLSupabaseJestDocker
Backend & Distributed

Distributed EV Charging System (Two-Phase Commit)

Concurrent distributed services in Go (goroutines, Gin, HTTP, and MQTT/Mosquitto) implementing the Two-Phase Commit (2PC) protocol (prepare, commit, abort) to coordinate distributed charging reservations across Docker containers.

GoGinTwo-Phase Commit (2PC)MQTTDocker
AI & ML

Neural Networks & Computational Intelligence from Scratch

Pure Python implementations of foundational ML models without black-box frameworks: Multilayer Perceptron (Backpropagation), Kohonen Self-Organizing Maps (SOM), Adaline (LMS Delta Rule), Rosenblatt Perceptron, Fuzzy Inference Controller, and Genetic Algorithms.

PythonMLP BackpropagationKohonen SOMAdaline LMSFuzzy LogicGenetic Algorithms
Luminus Project

Luminus Project

Academic dossier management system built as a RESTful API with Node.js and PostgreSQL, featuring secure stateless authentication and a Next.js frontend.

Next.jsNode.jsExpressPostgreSQLDocker
Minhas Compras BA

Minhas Compras BA

Web application for importing and analyzing SEFAZ-BA NFC-e electronic invoices with FastAPI, Python, React, and PostgreSQL.

FastAPIPythonReactPostgreSQLDocker