B.Tech CSE (Data Science) • Haldia Institute of Technology • Building ML Solutions for Tomorrow
Hello! I'm Srishti Kumari, a passionate Data Science and Machine Learning enthusiast currently pursuing my B.Tech in Computer Science and Engineering (Data Science) at Haldia Institute of Technology. My journey into the world of data began with curiosity about how algorithms can uncover patterns and drive intelligent decision-making.
With a strong foundation in Python, SQL, and Machine Learning frameworks, I specialize in building end-to-end ML solutions—from data preprocessing and exploratory analysis to model deployment. My experience spans predictive modeling, recommendation systems, and data visualization, with a focus on solving real-world problems through data-driven insights.
I'm particularly excited about applying ML to impactful domains like prediction systems, recommendation engines, and analytical dashboards. Whether it's engineering features for better model performance or crafting visualizations that tell compelling data stories, I thrive on transforming raw data into actionable intelligence.
To build impactful AI solutions that drive real-world value through predictive analytics, intelligent automation, and data-driven decision-making. I'm committed to continuous learning and staying at the forefront of ML innovation.
ML Projects
GitHub Commits
YGPA
Community Member
Organized by proficiency level and continuously expanding
Deep learning framework for neural networks and advanced ML models
Flexible deep learning library for research and production
ML model deployment and APIs
Demonstrated through end-to-end ML projects tackling real-world challenges
Public Relations Member at DSCH, leading outreach and event coordination
Active team player in technical clubs and group projects
Constantly expanding skill set in emerging ML/AI technologies
End-to-end Machine Learning solutions demonstrating technical depth and practical impact
Impact: Built a production-ready ML system achieving 92% R² score, enabling data-driven pricing decisions for automotive dealers and consumers
Problem Context: Accurate car valuation is critical for buyers, sellers, and dealers. This project implements a comprehensive ML pipeline to predict car prices based on features like brand, model year, mileage, fuel type, and transmission.
Handling non-linear relationships between features and price, dealing with multicollinearity, and ensuring model generalization across different car categories.
Problem: Binary classification challenge to predict passenger survival on the Titanic using demographic and ticket information.
Dataset: Kaggle Titanic dataset (891 training samples, 418 test samples)
Problem: Content-based recommendation system to suggest similar movies based on plot descriptions, genres, and keywords.
Dataset: TMDB 5000 Movie Dataset with metadata and credits
Problem: Comprehensive exploratory data analysis of COVID-19 trends across Indian states to identify patterns and insights.
Dataset: Daily COVID-19 cases, deaths, and recoveries by state (2020-2021)
YGPA: 7.87/10
Ongoing - Coursera
Completed
Completed
Completed
Leading outreach initiatives, organizing technical workshops, and coordinating events to promote data science awareness on campus. Responsible for community engagement and building connections with industry professionals.
Contributing to social welfare initiatives and community development programs while balancing technical pursuits with social responsibility.
Maintaining a consistent GitHub presence with 1000+ commits, showcasing ML projects and contributing to the data science community.
View GitHub ProfileWriting technical articles to share knowledge and explain complex ML concepts in accessible language (see Blog section).
I'm actively seeking internships and entry-level opportunities in Data Science and Machine Learning. Let's connect!
Haldia, West Bengal, India