Jovan Popovic

Background

About Jovan Popovic

I’m a Computer Science (AI & Data Analytics) Honours student at UTS, focused on turning advanced machine learning into practical tools that actually help people. My current thesis uses time-series forecasting (SARIMAX, XGBoost, LSTMs, Prophet) to support prepaid electricity customers in remote communities, building models and UI concepts that improve bill predictability, reduce disconnections, and tackle energy insecurity. Alongside that, I’ve worked on a privacy-aware multi-LLM routing system (LangGraph + local/cloud models with sensitivity classification), PII-masking with transformer-based NLP, and football analytics using computer vision and tracking for tactical insights.I enjoy bridging research and real-world impact: structuring messy data, building reproducible pipelines (Python, R, GitHub, MLflow, config-driven experimentation), and explaining results in a way non-technical stakeholders can actually use. I’m especially interested in roles at the intersection of AI engineering, data science, and product—where I can own models end-to-end, from problem framing through to deployment and UI integration.Previously, I’ve gained industry experience through internships at Alloy as a Junior Software Engineer, working on user experience and web building. Alongside that I have had experience as a Data Analyst at Ellipsis and Company working on frameworks to improve customer retention in customer loyalty programs.