👋 Hello! I'm Zairo Bastos, an AI Research Engineer and Master's student in Computer Science at the Federal University of Ceará (UFC). My research and professional work focus on Artificial Intelligence, Large Language Models (LLMs), Time-Series Forecasting, and Prompt Engineering.
🚀 I am passionate about applying advanced AI and data analysis to build innovative, scalable solutions for real-world challenges. Currently, I am bridging the gap between cutting-edge research and production by developing LLM-driven applications using microservices architecture for government and research institutions.
- Languages: Python, JavaScript, TypeScript
- AI, ML & Data: PyTorch, Pandas, Scikit-learn, Transformers (Hugging Face), LangChain, LlamaIndex, Pinecone
- LLMs: GPT-4/GPT-4o, Gemini 1.5, LLaMA 3, DeepSeek
- Web & Architecture: Microservices, Docker, ReactJS, Node.js, Streamlit, Flask
- DevOps & Tools: Git, Linux, Jupyter, Azure, Google Colab
🏛️ Casa Civil (State Government): Leading the development of an automated official document (ofícios) generation system utilizing LLMs, backed by a scalable Docker-based microservices architecture.🧠 Kunumi Research: Leading a research group focused on Prompt-Guided Forecasting, developing innovative methods to enhance the predictive capabilities of LLMs on complex time-series data.🤖 AI Agents (TCE-CE): Developed and deployed LLM-based solutions and intelligent agents to automate institutional FAQ and customer service (SAC), heavily optimizing prompt engineering.📈 Tax Anomaly Detection (SEFIN): Applied machine learning algorithms (Scikit-learn) and data visualization to detect anomalous behavior in corporate entities and analyze tax evasion.
📫 Feel free to explore my repositories or reach out for a chat! Email | LinkedIn | Google Scholar
👀 Profile Views