00.00.00

All Writing

Notes on AI and software engineering

Mixtral of Experts: Top-2 Routing Gives 47B Capacity at 13B Active Compute

LLaMA 2: How Three Borrowed Techniques Fit a 70B Model on Two GPUs

Attention Is All You Need: What the Paper's Heads Are Actually Doing at Each Layer

Beyond try/except: Architecting Robust Error Handling in Python Applications

Demystifying Reinforcement Learning: A Beginner's Guide to the Math

Exploration vs. Exploitation: A Deep Dive into Multi-armed Bandits

Reinforcement Learning: First Principles