Blog
Learnings, reflections, paper summaries, and notes.
Outside the Agentic Loop
Everyone in the valley reduces an AI agent to a while loop. The loop is the easy five percent. This is a plain map of the other ninety five, the part that actually makes an agent good.
All (Machine) Learning is just a measure of surprise
Hamming's back-of-envelope math made me realize my own math fundamentals weren't strong enough, so I rebuilt it from the floor (exponentials, logs, surprise) until I got to cross-entropy loss, finally explaining itself from the bottom up.
A Month Off Code
Reflections from a month in San Francisco where I took a deliberate break from coding agents, two hackathons, and what stepping back taught me about what I want to build.
Build something useful
Modernizing a Dental Practice
Cli Is Cool Again
How we are moving away from GUIs with the rise of Coding Agents
[Learnings] 001 - AI Applied (Prompt Engineering and Evaluation)
The two main ways AI Engineers boost their outputs — Prompt Engineering and Evaluation.
[Reflections] 001 - The Thinking Game
Watching The Thinking Game about DeepMind and processing where humans fit in the age of AI.
[Learnings] 000 - AI Applied (Foundations)
A practical guide to working with AI models — API basics, tokens, embeddings, and system prompts.
[Reflections] 000 - Reset
Getting back to writing. Building. Doing.
[Note to Self]
A quick note about getting back to writing and learning.
[PapersSummarized] Deep Residual Learning for Image Recognition (2015)
K. He et al. — How residual connections solved the degradation problem in deep networks. Rating: 3/5.
[PapersSummarized] Very Deep Convolutional Networks For Large-Scale Image Recognition (2015)
K. Simonyan et al. — VGG paper showing deeper is better with smaller filters. Rating: 1/5.
[PapersSummarized] ImageNet Classification with Deep Convolutional Neural Networks (2012)
A. Krizhevsky et al. — The paper that lit the deep learning revolution. A must read. Rating: 5/5.
[PapersSummarized] Backpropagation Applied to Handwritten Zip Code Recognition (1989)
Y. LeCun et al. — The first successful large-scale commercial application of backpropagation. Rating: 3/5.
[NotesToSelf] Intuition Behind Why ReLU Works So Well
Breaking down why ReLU is the go-to activation function in neural networks.
[PapersSummarized] Learning Representations by Back-Propagating Errors (1986)
G. Hinton et al. — The seminal paper that laid the foundation for all deep learning. Rating: 2/5.
[PapersSummarized] A Few Useful Things to Know about Machine Learning (2012)
P. Domingos — 12 practical insights every ML practitioner should know.
[PapersApplied] Mining Association Rules between Sets of Items in Large Databases (1993)
R. Agrawal et al. — The Apriori Algorithm for finding item relationships in transaction data.
[PapersApplied] Induction of Decision Trees (1986)
J. R. Quinlan — The precursor to C4.5 and C5 algorithms for creating decision trees.