João Victor Castro

Machine Learning engineer and AI researcher with hands-on experience in data analysis, classical ML, and LLM security. Currently pursuing a Bachelor's degree in Artificial Intelligence at the Federal University of Goiás (UFG) — one of Brazil's leading AI programs — while conducting research at CEIA (Center of Excellence in Artificial Intelligence), a top-tier applied AI institute.

I help companies turn raw data into actionable decisions and build ML-powered solutions grounded in both academic rigor and real-world applicability. My work spans exploratory data analysis,

predictive modeling, BI dashboards, full-stack AI application development, and AI safety research — with published results in peer-reviewed venues. Most recently, I built a production-ready

fact-checking chatbot for UFG’s dAurora project (in partnership with Anatel and the Goiás Electoral Court), integrating generative AI, custom security layers, and cloud deployment from the round up.

What I deliver:

• Predictive ML models (classification, clustering, regression) applied to real business problems

• Data analysis pipelines using Python, SQL, and Looker Studio — from raw data to actionable

dashboards

• LLM security evaluations and AI safety assessments

• Clear communication of technical results to non-technical stakeholders