Numerical Sensitivity: Why FP16 Breaks NAdam and How to Fix It
A visual deep dive into floating-point arithmetic, numerical stability, and...
Teaching Machine to be Smart
Abhik Sarkar is an experienced Python Machine Learning Engineer with over eight years of expertise in scalable, intelligent solutions. He currently leads the Machine Learning department at a high-growth startup specializing in computer vision technologies for safety and security.
Articles
SAM's Multi-Mask Ambiguity: A Visual Deep Dive
Deep dive into how SAM resolves point prompt ambiguity through...
YOLOv11 Loss Functions Explained: Interactive Visual Guide
Understand YOLOv11's loss functions through interactive visualizations. Compare IoU variants...
H.264 Implementation & Real-World Applications: From Theory to Practice (Part 3 of 3)
Part 3 of the H.264 series. Explore practical implementation aspects:...
Work
- Company
- Cloudastructure Inc
- Role
- Director, Machine Learning
- Date
- Company
- Quantiphi Inc
- Role
- Machine Learning Engineer
- Date
- Company
- Deloitte
- Role
- Business Technology Analyst
- Date
Bookmarks
Resources I find valuable and recommend to anyone interested in ML and engineering.
BRRR: A faster Python
Horace He
The Unreasonable Effectiveness of Transformer Models
Andrej Karpathy
Transformers from Scratch
Brandon Rohrer