Designing Machine Learning Systems
The best end-to-end guide on taking ML from notebook to production. Changed how I think about data distribution shifts and monitoring.
Reading List
Books, papers, and resources that have shaped how I think about ML systems, product building, and the real-world problems they solve.
🧠 ML Research
Attention Is All You Need
The transformer paper that reshaped NLP and, eventually, everything else. I keep returning to the multi-head attention diagrams.
Deep Learning
Dense but indispensable — the chapters on optimisation and regularisation are foundational for anyone building neural nets.
🚀 Product & Startups
The Lean Startup
Build-measure-learn loops applied directly to how I approached early Pulp AI iterations. Lean validation saved us months.
Zero to One
Contrarian thinking on monopoly vs competition. The question 'What important truth do very few people agree with you on?' stays with me.
Inspired
Practical product management frameworks. The distinction between discovery and delivery shaped how I run product experiments.
🏥 Public Health
Guns, Germs, and Steel
A sweeping look at how geography and biology shaped civilisation — informed my perspective on health infrastructure disparities.
The Emperor of All Maladies
A biography of cancer that reads like a thriller. Deepened my appreciation for how diagnostics and public health intersect.
📊 Economics
An Introduction to the Analysis of Algorithms
Rigorous yet readable. The probabilistic analysis chapters gave me new tools for reasoning about system performance at scale.
Poor Economics
RCT-driven development economics. Directly inspired the evaluation methodology we used in the TB adherence project.