Maulik Mathur

I’m a data science intern and AI/DS student who builds end-to-end analytics and machine learning workflows. I’m proficient in Python, SQL, machine learning, deep learning, and Big Data technologies, and I enjoy turning messy data into clear decisions.

In my Power BI work, I designed an end-to-end business intelligence pipeline using the AdventureWorks dataset, including Power Query transformations, a fact/dimension model, and DAX-driven executive dashboards. I emphasize interactive reporting with drill-through navigation, bookmarks, slicers, and AI visuals like Key Influencers and Decomposition Tree to help stakeholders track sales performance, profitability, and customer trends.

I also work with real-time and research-grade pipelines—fetching Indian stock market data with the Upstox API, streaming it via Apache Kafka, processing time-series with Apache Spark Structured Streaming, and presenting results through Flask dashboards. For medical imaging, I implemented Simple U-Net and Attention U-Net segmentation and evaluated federated learning against centralized training using Dice Coefficient and Jaccard Similarity, while exploring ways to reduce message exchange rounds.