Hi, I am Shashank Goel
An aspiring data science student who's passionate about Machine Learning, Finance & real-world impact
About Me
I'm a data science student with a deep curiosity for how data can transform industries and create meaningful impact in the real world. My journey in tech is driven by a passion for understanding complex problems and solving them through innovative approaches.
With a focus on machine learning and finance, I'm constantly exploring the intersection of technology and practical applications. I believe in building solutions that not only work technically but also make a tangible difference in people's lives.
When I'm not coding or analyzing data, you'll find me staying updated with the latest developments in AI, learning some new concept in Finance or exploring new ways to apply data science to solve real-world challenges.
Interests
- ▹Machine Learning & AI
- ▹Data Analysis & Visualization
- ▹Finance
- ▹Learning something new
Currently
- ▹Pursuing B.Tech Data Science
- ▹Building ML Projects
- ▹Learning New Technologies
Featured Projects
A collection of projects showcasing my technical expertise
INR Exchange Rate Prediction
• Developed comprehensive monthly time-series dataset spanning 25+ years (2000-2025) by integrating multiple data sources including global commodity prices (crude oil, gold, silver), international market indices (S&P 500, NASDAQ, SENSEX), and geopolitical event indicators, engineering 40+ features with lag variables and rolling statistics
• Implemented and optimized 10+ forecasting models including deep learning architectures (LSTM, GRU), gradient boosting algorithms (XGBoost, LightGBM), ensemble methods (Random Forest), and statistical approaches (ARIMA/SARIMA, ETS), applying hyperparameter tuning, regularization techniques, and cross-validation to maximize predictive accuracy
Vehicle Type Prediction
• Developed a full-stack vehicle classification app using TensorFlow CNN, FastAPI, and Next.js, delivering real-time image predictions
• Integrated FastAPI RESTful API with a Next.js frontend, enabling seamless vehicle image uploads and accurate predictions
• Optimized image processing and deployment pipeline with Python, TensorFlow, FastAPI, and Next.js for high-performance results
Department of Government Efficiency Web-Scraper
• Engineered an automated web scraping solution to collect public project data, specifically leases, grants, and contracts, from over 2,000 US Government web pages
• Structured and cleaned the extracted data, reducing data preparation time by 10%, then created interactive dashboards in Tableau to highlight departmental savings and key inefficiencies
Backpack Price Prediction
• Developed predictive models (Random Forest, Neural Network, XGBoost, LightGBM, CatBoost) to forecast backpack prices based on product attributes
• Improved accuracy through feature engineering, optimizing model performance using RMSE and R² metrics
Indian Birds Classification Model
• Designed and implemented Convolutional Neural Network (CNN) and K-Nearest Neighbors (KNN) models to classify over 25 bird species
• Leveraged advanced data augmentation techniques to increase model robustness and improve classification accuracy to 70%
Text Prediction & Information Retrieval Assistant
• Implemented contextual text prediction and RAG (Retrieval-Augmented Generation) models for enhanced AI-driven user interactions and seamless response generation
• Enabled efficient information retrieval and accurate text prediction, optimizing user experience
Skills & Technologies
A comprehensive toolkit I've developed through projects and continuous learning