I build and document systems in machine learning, AI, and product engineering.
This site is my working knowledge base — where ideas are tested, refined, and connected.
⭐ Featured Series
A 11-part deep dive into one of the most important ideas in modern AI.
The series starts with the limitations of traditional sequence models, builds intuition for attention, and gradually progresses through self-attention, multi-head attention, positional encoding, and the Transformer architecture.
Rather than focusing only on formulas, the goal is to understand why attention changed AI and how the ideas fit together as a system.
🚀 AI Systems Thinking (Start Here)
A structured deep dive into how modern AI systems actually work:
from prompts → workflows → agents → production systems.
This is not a collection of tips.
It is a connected system for understanding how to design reliable AI applications.
Featured
- What 10-fold Stratified Cross-Validation Really Does
- Batch Normalization Explained Simply
- CIFAR-10 Image Classification — Build vs Buy
Core Topics
How to use this site
- Start with AI Systems Thinking if you’re interested in applied AI and systems design
- Explore Projects to see real implementations
- Use Cheatsheets for quick recall
- Dive into Notes for deeper understanding
What this site is
- A place to think in public
- A system for connecting ideas
- A record of building, not just learning
I learn by building, testing, and explaining — this site reflects that process.