
Gradient Boosting vs. Neural Networks: When Classical ML Still Wins
A deep-dive into why gradient-boosted trees consistently outperform deep learning on tabular data.
Read MoreAI/ML and Cybersecurity enthusiast building intelligent, secure systems. I love blending Deep Learning with domains like security, space science, and real-time systems—and I also enjoy Competitive Programming to sharpen my problem-solving.
Who I am, what I do, and what I’m great at
I’m Debnil Pal, a creator who believes technology should not only solve problems but also inspire change. My work lies at the crossroads of AI, cybersecurity, and innovation, where creativity meets logic. I build tools, design systems, and explore research that empower humans to work smarter with technology. The future is intelligent — and I’m coding my way into it.
AI-first skillset spanning ML/DL, Security, and Algorithms
Selected work from my GitHub
Compliance knowledge and sourcing tool built for a hackathon.
AI-powered risk scoring using alternative financial signals.
Boulder and landslide detection experiments on lunar imagery.
Deep dives into ML, NLP, Deep Learning, and Quantum Computing

A deep-dive into why gradient-boosted trees consistently outperform deep learning on tabular data.
Read More
How tokenization strategies silently shape every NLP model's performance, from BPE to byte-level.
Read More
Why residual connections are the single most important architectural innovation in deep learning.
Read MoreLet's discuss your next project
+91 9933182503
India, West Bengal, Kolkata
Have recommendations or want to discuss interesting topics? Join the conversation!
Join Discussion