Talha Khalid

I’m a B.Sc. Computer Science & Information Technology student specializing in Machine Learning and Data Engineering, focused on turning messy data into reliable, production-ready decisions. I enjoy designing predictive models, ETL pipelines, data warehouses, and retrieval-augmented LLM systems that perform under real constraints.

In my internships, I designed and deployed ML pipelines across multiple datasets, improving model accuracy by 12% on average using XGBoost with cross-validation and hyperparameter tuning. I also reduced overfitting through stratified sampling and feature importance pruning, and delivered analytics used by 3+ departments for planning.

I build scalable data products and automated workflows—like centralized KPI/ROI tracking used by 4 internal teams and reporting pipelines that cut manual effort by 35%. On the web side, I’ve integrated React frontends with Flask REST APIs and optimized backend queries to improve response time by 25%.

As a project lead for a RAG-based AI Health Assistant, I improved contextual relevance by 25% using optimized embedding indexing and semantic retrieval, while reducing hallucinations through structured chunking and retrieval evaluation. I’m motivated by measurable impact, clean data practices, and cloud-ready engineering.