Skip to content
View FelipeAraujoBS's full-sized avatar
12 followers · 9 following

Achievements

Achievement: YOLOAchievement: Quickdraw

Achievements

Achievement: YOLOAchievement: Quickdraw

Block or report FelipeAraujoBS

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
FelipeAraujoBS/README.md

Felipe de Araújo

Backend Engineer · Distributed Systems · AI Engineering

LinkedIn Dev.to Email

Backend Engineer specializing in distributed systems and AI Engineering. I build production-ready pipelines, from event-driven architectures to RAG systems with hybrid retrieval, reranking, and LLM-as-judge evaluation. 2+ years of professional experience building and operating backend systems in production, with direct impact on 3,000+ users. Open to Backend / AI Engineering roles (mid-level), remote-first.

Background in Geophysics, which grounds my applied math foundations for ML.

Currently finishing a degree in Software Engineer at Universidade Católica do Salvador (expected graduation: 2028);

🔨 Featured Projects

Uma Busca de Gelo e Fogo — Full production system of 3 microservices over the A Song of Ice and Fire corpus. Backend: full-text search engine (SQLite FTS5) over 10 books / 2,400+ chapters, with an EPUB parsing pipeline (POV detection, alias normalization) and a Fastify REST API. RAG Service: hybrid retrieval (BAAI/bge-m3 dense + BM25 sparse + RRF), cross-encoder reranking, and LLM-as-judge evals over ~66k paragraphs. Frontend: Next.js. ~250 daily accesses in production. TypeScript · Python · Fastify · SQLite FTS5 · FastAPI · ChromaDB · Groq · Next.js

Weather Monitoring System — Full-stack distributed weather monitoring pipeline: Python collector → RabbitMQ → Go worker → NestJS API → React dashboard. Real-time data, AI-generated insights, and CSV/XLSX export. Python · Go · NestJS · RabbitMQ · React · MongoDB

Logflow — Centralized real-time log ingestion and monitoring platform. Ingests logs via HTTP and gRPC, processes them asynchronously with BullMQ/Redis, and persists to MongoDB, with an Angular dashboard for filtering and analysis. Node.js · TypeScript · Fastify · gRPC · BullMQ · Angular TypeScript · React

Programming Languages:

Typescript Python Go


Frameworks & Libraries:

Node NestJS Fastify FastAPI React NextJS


Messaging & Infrastructure:

RabbitMQ Redis gRPC Docker


Databases:

Postgresql MongoDB SQLite


Other:

Git Linux Bash Swagger Prisma


Stats:

Felipe's Top Languages GitHub Streak

Currently:

  • Deepening expertise in agentic RAG architectures and LangGraph
  • Strengthening the applied math layer behind ML (Linear Algebra, Statistics, Probability), building on a Geophysics background
  • Writing about the process at dev.to/felipearaujobs

📧 felipearaujobs@hotmail.com
🌐 linkedin.com/in/felipe-de-araujo-b87386231

Pinned Loading

  1. A-Procura-de-Gelo-e-Fogo-Backend A-Procura-de-Gelo-e-Fogo-Backend Public

    Backend of the A Procura de Gelo e Fogo project. This backend consists of both the book parsing logic (built in Python), the REST API that serves the data, and the search logic (TypeScript and bett…

    TypeScript

  2. A-Procura-de-Gelo-e-Fogo-Frontend A-Procura-de-Gelo-e-Fogo-Frontend Public

    Frontend of the "A Procura de Gelo e Fogo" project, built with TypeScript, Next.js, and SQLite.

    TypeScript

  3. Uma-RAG-de-Gelo-e-Fogo Uma-RAG-de-Gelo-e-Fogo Public

    Uma-RAG-de-Gelo-e-Fogo microservice for the A Procura de Gelo e Fogo ecosystem. Combines hybrid retrieval (dense + sparse + reranking) with LLM generation to answer natural language questions about…

    Python

  4. gemini-chromadb-RAG-pipeline gemini-chromadb-RAG-pipeline Public

    A minimal RAG pipeline built from scratch in Python — no LangChain, no abstractions. Embeddings, vector search and LLM generation made explicit.

    Python

  5. weather-monitoring-system weather-monitoring-system Public

    Monitoramento do clima FullStack - Python collector → RabbitMQ → Go worker → NestJS API → React dashboard. Dados em tempo real + insights de IA + Exportação de arquivos CSV/XLSX.

    TypeScript

  6. logs-platform-logflow logs-platform-logflow Public

    Plataforma centralizada de ingestão e monitoramento de logs em tempo real. Recebe logs via HTTP e gRPC, processa de forma assíncrona com BullMQ e persiste no MongoDB — com dashboard Angular para fi…

    TypeScript 1