Prathmesh Bunde

End-to-end seismic data engineering & analytics platform on Databricks using Delta Lake and PySpark

End-to-end Customer Churn Prediction system showcasing real-world ML skills. Built with XGBoost, SMOTE, and Streamlit, featuring a deployable UI, probability-based risk insights, and production-ready architecture. Demonstrates modeling judgment, deployment, and business impact.

Predict Microsoft stock prices using an LSTM neural network in Python. The project includes data exploration, visualization (open/close prices, volume, correlations), preprocessing, LSTM model training, and predictions. Plots are saved for analysis. Built with TensorFlow, Pandas, NumPy, Matplotlib, and Seaborn.

No description.

Spring Boot backend system implementing secure REST APIs for e-commerce operations including product inventory management, order processing, and role-based authentication with real-world business logic

Credit Risk Prediction System is an end-to-end machine learning project that predicts loan default risk using customer financial data. It applies EDA, feature encoding, and advanced models like Random Forest and XGBoost, and is deployed via Streamlit for real-time credit risk assessment.

Data Structures and Algorithms - Daily Problem Solving

Research-grade quantitative finance engine built in Python with vectorized Monte Carlo simulation, stochastic asset modeling, option pricing, portfolio risk analytics, Sobol QMC, antithetic variates, interactive GUI support, and production-oriented software architecture

Repository for advanced Java development and object-oriented systems engineering at MIT-ADT University. Contains structured daily assignments, robust OOP implementations, and console-based portfolio applications detailing core concepts in software architecture, and memory optimization.

No description.

🎨 Enhance your GitHub profile with this amazing tool, which allows you to create a personalized README quickly and easily. This profile README generator is the best you'll find 🚀

10.7M-parameter decoder-only transformer trained from scratch on Tiny Shakespeare. Features an optimized Causal KV-Caching inference engine reducing generation runtime complexity to $O(T)$, dynamic shape tracking to eliminate asynchronous cuBLAS memory faults, mixed-precision acceleration, and a robust out-of-vocabulary validation guardrail.

PyTorch implementation of a Selective State Space Model (SSM) using the Mamba architecture. Designed as a diagnostic framework, it features a custom signal-to-noise synthetic dataset generator to evaluate latent state memory trajectories.

Research-grade distributed training engine for financial LLMs with CUDA custom kernels, PyTorch Lightning, and FSDP FULL_SHARD optimization.

No description.

No description.

No description.

No description.

No description.

Medical Chatbot with LLMs, LangChain, Pinecone, Flask & AWS

[](https://holopin.io/@sparkydev007)

A collection of my Google Colab notebooks documenting hands-on experiments in Machine Learning, Deep Learning, and Data Analysis. Includes model training, evaluation, and visualizations.

Predicts student marks using Linear Regression

Real-time M&A, IPO & Market Tracker

No description.

No description.

No description.

No description.

No description.

Dice C++ random roll

🚀✨ Help beginners to contribute to open source projects

Self online documentation of DSA learning journey

creating a decent looking website with parallax effect

a generic Login page use html and css

my first if else statement practice

A game of Flappy Bird made in pure JavaScript, HTML and CSS

No description.