- LLMs
- transformers
- MLOps
- NLP
- deep-learning
- systems
- deep-learning
- mlops
- ai-research
- systems
•
•
•
•
•
•
•
•
•
-
Superposition: How Neural Networks Secretly Cram More Concepts Than They Have Neurons
Most engineers assume one neuron = one concept. They're wrong. Neural networks pack thousands of overlapping features into far fewer neurons using a trick from high-dimensional geometry — and understanding this changes everything about how we interpret AI.
-
Understanding Transformers: The Architecture Behind Modern AI
A deep dive into the transformer architecture — the building block of GPT, BERT, and virtually every modern AI system. We'll implement multi-head attention from scratch.
-
From ARIMA to LSTM: Choosing the Right Model for Time-Series Forecasting
A structured beginner-to-advanced guide to understanding ARIMA and LSTM models, how they work mathematically, and when to use each in real-world forecasting systems.