Data Science and Machine learning Intern

Hankernest Technologies Pvt Ltd is an EdTech startup dedicated to transforming education by providing schools and colleges with technology labs, training programs, and industry-focused workshops.

• Designed and deployed a crime investigation game using Python, Streamlit, and SQL, analyzing over 900,000 criminal incidents to uncover patterns, enhancing the game’s decision-making accuracy by 15%.
• Scraped data for 10,000+ mobile products from Amazon using Python and BeautifulSoup, exporting results to Excel for a detailed market trends report. Also, built a house price prediction model using Linear Regression, improving prediction accuracy by 20% through optimization of data preprocessing and feature scaling techniques.

R&D Intern

mFilterIt is a digital brand protection and ad-fraud prevention company that helps global brands ensure safe and transparent marketing.

• Engineered a serverless web scraping pipeline using AWS Lambda and API Gateway to extract URLs, images, videos, GIFs, JavaScript, and HTML tags for scalable automated data collection, incorporating a company-provided proxy system to bypass site restrictions and anti-scraping mechanisms.
• Designed an AWS Lambda function to classify website themes via API using Grog and LLaMA models, incorporating JWT token- based authentication (H3S25 algorithm) and AWS S3 integration for secure data storage.
• Implemented two YOLO-based logo detection models (37 classes and 14 classes), performing full pipeline work including annotation with LabelImg, dataset augmentation with Roboflow, and training on AWS EC2 GPU instances using high-quality images; achieved over 90% detection accuracy and improved brand monitoring performance.
• Performed EDA on e-Commerce data to uncover product-level insights and created lip movement detection using MediaPipe.
• Built a thread-safe GPT token calculator that automatically tracks API usage and logs daily summaries in CSV format. Implemented thread locking to ensure accurate logging during concurrent requests, enabling usage pattern analysis and cost optimization.

Bachelor of Technology

CGPA: 7.0

Higher Secondary School

Percentage: 95%