Sohan Dipesh

About Sohan Dipesh

As a Data/Business Analyst and software enthusiast with hands-on experience in Python, SQL, Power BI, and Machine Learning, I enjoy solving real-world problems through data-driven insights and clean, user-centric software. Currently pursuing my B.Tech in Electrical and Electronics Engineering at VIT Chennai (CGPA 8.97), I’ve applied analytical and technical skills across diverse projects. I built a credit default prediction model with 92% accuracy, developed a Power BI dashboard that improved revenue visibility for a hotel chain by 25%, and performed EDA using Python and Excel to drive business insights. As a Research Intern at the Centre for Neuroinformatics (CNI), I engineered stress-based EEG features and automated signal preprocessing workflows, boosting seizure classification performance by 14% and reducing manual effort by 40%. On the development side, I built and deployed a fully responsive travel website using ReactJS, achieving 70% faster load times and 40–50% improvements in accessibility and SEO using Lighthouse audits — showcasing my interest in creating fast, functional user interfaces. My technical toolkit includes Python (Pandas, NumPy, Matplotlib), SQL, Power BI, Excel, Scikit-learn, ReactJS, and AWS. I also hold certifications from Google (Data Analytics) and Microsoft (Azure Data Fundamentals – DP-900). I’m currently seeking Data Analyst, Business Analyst, or data-driven software roles where I can apply my hybrid skills to build smart, scalable solutions. Let’s connect!

Experience

1. Research Intern

   Vellore Institute of Technology

   Jun 2025 — Present · Chennai, IN

   Research InternCentre for Neuroinformatics (CNI), VIT ChennaiMay 2025 – Jul 2025Improved seizure classification accuracy by 8–14% by engineering EEG features using stress-based physiological ratios such as Beta/Alpha and Theta/Beta.Built and evaluated machine learning models including Random Forest, SVM, KNN, Logistic Regression, and LSTM to compare baseline vs. stress-enhanced EEG datasets.Automated EEG signal preprocessing and feature extraction using Python, reducing manual effort by 40% and enabling scalable experimentation.Designed a reproducible ML pipeline to support consistent evaluation across patients and models, streamlining research for future clinical applications.

Education

• Kendriya Vidyalaya

  92%

  2009 — 2019

• Kendriya Vidyalaya

  PCMB

  2019 — 2021

• Vellore Institute of Technology

  Bachelor of Technology - BTech

  2022 — 2026