- I'm Brasilian/American
- 31 years old
- Training to join the US Military
- I specialize in creating applications with intelligent, real-time search. These apps utilize hybrid search—a combination of semantic and keyword methods—to understand user intent. The underlying vector embeddings ensure contextual accuracy, while WebSockets provide the scalable, low-latency infrastructure for instant data retrieval
- Implementing local LLM solutions with Ollama.
- Developing with Vite + React for fast, scalable web apps.
- React, Vite, Svelte, Blazer
- Tailwind CSS, Bootstrap
- JavaScript/TypeScript
- Node.js
- Python (Flask, Django, FastAPI), C# (.NET)
- API Development, Bash Automation
- Vector Processing, RAG Architecture
- Search Systems, LLM Integration
- Bert op | Install this "ollama pull snowflake-arctic-embed"
- PostgreSQL, MongoDB, SQLite, SQL
- Docker, PG VectorDB
- Vector (Faiss, JSON for cluster)
I follow a structured problem-solving process inspired by George Pólya:
- Analyze: Understand objectives and constraints.
- Design: Create scalable, efficient solutions.
- Implement: Write clean, maintainable code.
- Optimize: Refine systems and scale for growth.
- Built with Python, Vite + React, and Ollama LLM.
- Features semantic search using vector embeddings.
- View Project →

