skip to content

Search

Books

Long-form guides to how digital advertising actually works and makes money, built from first principles and from the data side of a DSP and ad exchange. Read free, chapter by chapter.

Complete · 25 chapters

The Monetization Leader's Playbook

A Leader's Guide to Scaling Digital Advertising

A sibling to The Ad Dollar's Journey, one altitude up and inside the walled gardens. Written for the platform ads leader — the VP/GM of Ads & Monetization at a major platform — it teaches the strategic calls a leader must make and defend: ad load, the auction, automation, measurement, privacy, and the agentic future. Every mechanic is taught only in service of a decision.

Read the book →
Complete · 21 chapters

From Noise to Image

A Chronological Field Guide to AI Image Generation

A dated, cited reference history of machines that make pictures: from the first variational autoencoders and the adversarial game of GANs, through the diffusion revolution, latent diffusion and Stable Diffusion, the control and fine-tuning ecosystem, to flow matching and today's frontier. Every chapter carries the original mathematics with plain-English translations, the original papers, and a product lens on why each breakthrough moved markets.

Read the book →
Complete · 18 chapters

From Frames to Worlds

AI Video Generation, from GAN Clips to World Models

Video is the hardest canvas: every pixel must agree with the last frame. This book traces generated video chronologically — early GAN clips, autoregressive tokens, video diffusion, the Sora moment, the commercial race, and the turn toward interactive world models — with the math, the papers, and the economics of the most compute-hungry medium in AI.

Read the book →
Complete · 22 chapters

The Language Machines

From word2vec to Reasoning Agents: A Chronicle of Foundation Models

The spine of the modern AI story: words become vectors, attention becomes the transformer, scale becomes a law, feedback becomes alignment, and chat becomes reasoning and agents. An encyclopedic, paper-by-paper chronicle of foundation models with every core equation explained twice — once in LaTeX, once in plain English — and a PM lens on how each step became a business.

Read the book →
Complete · 20 chapters

The Engine Room

Hardware and Training Methods: How Silicon Made Modern AI Possible

Model history is hardware history. This book tells the story the other three depend on: how GPUs, TPUs, precision formats, parallelism, memory tricks, LoRA, quantization, MoE and inference engines successively raised the ceiling on what could be trained and served — culminating in the co-evolution thesis: every model breakthrough rode a silicon breakthrough, and vice versa.

Read the book →
Complete · 24 chapters

The Google Marketing Stack

How One Company Built Both Sides of the Table

An architecture deep-dive on the tools Google actually ships to advertisers, publishers, and marketers — Search Console, GA4, Google Ads, Search Ads 360, DV360, Campaign Manager 360, Google Ad Manager, AdX, Tag Manager, Merchant Center, and Data Studio. Not a how-to-use-each-product manual — a study of how they were assembled, why they're shaped the way they are, and how they wire together into one system that clips a fee on nearly every impression on the web.

Read the book →