Imad Dabbura
Blog
Paper Summaries
TIL
Projects
DL Tips & Tricks
More
Library
Notes
Data Engineering
About
Imad Dabbura
AI Scientist
Understanding AI by building it from scratch — long-form posts, paper summaries, and open-source projects from first principles.
Learn more about me →
Start Here
Make ML Systems Ship Again
ML Systems
Hard-Learned Lessons in Shipping Software (AI/ML) Projects
Machine Learning
Deep Learning
Software Engineering
Building GPT(2/3) from Scratch: Turning Theory into a Working Transformer
NLP
I Built My Own PyTorch (Tiny Version) — Here’s Everything I Learned
ML Systems
The Transformer Architecture: A Deep Dive
NLP
Inside LSTMs: Implementing and Optimizing Sequential Models from First Principles
NLP
No matching items
Latest Writing
Why Cross-Entropy Loss Can Never Reach Zero
Why softmax can’t reach the corners of the simplex, and what that does to the predictor
Machine Learning
Deep Learning
Optimization
Apr 27, 2026
10 min
Why Autograd Needs Floats, Not Ints
Calculus is continuous — and integer tensors quietly kill the gradient signal
Deep Learning
Machine Learning
Apr 21, 2026
11 min
Make ML Systems Ship Again
A practitioner’s guide to finding and fixing the one bottleneck that governs your system’s performance.
ML Systems
Sep 21, 2025
22 min
No matching items
Back to top