Hi, I’m Prabal completed MEng Computer Science at the University of Birmingham. I’m passionate about building useful products with deep learning and low level programming, and exploring how AI can be applied in real-world scenarios like finance and Media.
- Developing TorchPINN, a production-ready PyTorch library for solving partial differential equations using Physics Informed Neural Networks built from the ground up with modular design and research-grade scalability.
- Building a multi-modal model to classify hate speech in video content using text, audio, and visual signals.
- Creating a minimal and useful college application assistant that helps students shortlist universities, track deadlines, and organize application documents, all in one place.
- Exploring LLMs, cloud deployment, and advanced deep learning training pipelines.
- Always tinkering with side projects in finance, machine learning, and full-stack development.
- Deep learning, especially neural architecture design and interpretability
- Financial modeling and algorithmic trading
- Scalable backend systems and dev workflows /bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)" brew install node node -v npm -v git clone cd wingmanio cd /path/to/wingmanio chmod +x ./create-desktop-dmg.sh ./create-desktop-dmg.sh open desktop/dist
I ranked 7999 out of 1.5 million in JEE Advanced — one of the world’s toughest exams. It taught me how to enjoy solving difficult problems.
- Email: yadavsk731@gmail.com
- LeetCode
Languages: Python, C, SQL
Frameworks: PyTorch, TensorFlow, Flask, React, Node.js
DevOps: Docker, Git, Linux
ML/AI: HuggingFace, OpenCV, scikit-learn, MLflow
-
Multi-Modal Hate Detection
AI model for detecting hate speech in videos using text, audio, and frames. -
MoneyMaven
Full-stack finance tracker with charts, budgeting tools, and CI/CD pipeline. -
Forex Analysis
Built predictive models and analysis tools for trading FX pairs.
Thanks for dropping by! Always open to collaboration and learning from others.