Matt Smith

Welcome

I've spent my career navigating the machinery of search and growth for some of the biggest names in tech—from Dell and Edmunds to my current role at RVO Health. But the most exciting part of my day isn't always the 'Win'; it's the Spark.

This blog is my digital workbench. It's where I store my projects, document my experiments with LLMs, and share the leadership lessons I pick up while managing high-scale teams.

What you'll find here:

  • The Build: A look under the hood of my side projects and AI workflows.
  • The Strategy: Practical takes on SEO, paid marketing, and the evolving AI landscape.
  • The Human Side: Leadership, 'dad life' moments, and the occasional sci-fi-fueled daydream.
I believe the best way to predict the future of AI is to start building with it. This is me doing exactly that—messing around, figuring it out, and sharing the results in real-time.

Matt Smith

A Google engineer was charged with making $1.2M on Polymarket using confidential "Year in Search" data. We pulled his complete on-chain trading record — and the insider edge is sitting in public for anyone who can read it.

On May 27, 2026, the Department of Justice charged a Google engineer, Michele Spagnuolo, with making more than $1.2 million on Polymarket as the user "AlphaRaccoon," allegedly by trading on confidential Google data. The charge made headlines for a day and moved on. The part worth a second look isn't the charge. It's that the entire alleged crime has been sitting in public the whole time, on-chain and fully reconstructable, for anyone who knows how to read it. Polymarket settles on Polygon, so every bet AlphaRaccoon placed — the price he paid, the day he placed it, what it eventually paid out — is permanently visible. So we pulled his complete trading record. What he actually knew Every December, Google publishes its "Year in Search" — the rankings of the most-searched people, shows, and topics of the year. Polymarket runs prediction markets on the outcome: "Will [X] be the #1...

Featured
16 min read
#Polymarket#Insider Trading#Data Viz

An interactive visual guide to understanding paid search advertising on Google and Meta platforms.

Interactive Guide to Paid Search on Google and Meta This interactive tool walks you through the fundamentals of paid advertising on Google Ads and Meta Ads — covering campaign structure, ad auctions, quality signals, targeting, tracking pixels, key metrics, and bidding strategies. Pick a platform and explore. PAID ADVERTISING LEARNING LAB Each path covers structure, auction, targeting, pixels, metrics & bidding. G Google Ads Search · Display · Shopping M Meta Ads Facebook · Instagram · Reels You can switch platforms any time from inside the guide. G Google Ads M Meta Ads ← Back

Featured
17 min read
#Paid Search#Google Ads#Meta Ads

Claude just launched interactive charts and diagrams in beta — I took it for a spin by visualizing S&P 500, Dow Jones, and federal deficit data by presidency.

Economy Performance by President: Testing Claude's New Interactive Charts Anthropic just launched a new beta feature in Claude — interactive charts and diagrams. Instead of static images or markdown tables, Claude can now generate fully functional, self-contained HTML dashboards with real interactivity. Tabs, tooltips, responsive layouts, dark mode — the works. I figured the best way to test it was to throw a real question at it: how have the S&P 500, Dow Jones, and federal deficit changed across every U.S. presidency in my lifetime? One prompt. One self-contained HTML file. No frameworks to install, no build step — just Chart.js loaded from a CDN and clean vanilla JavaScript. The result speaks for itself: Presidential economics dashboard S&P 500, Dow Jones, and federal deficit changes across every U.S. presidency since 1981. S&P 500 performance Dow Jones performance Federal deficit change Democrat Republican Trump II (partial term) Click through the tabs...

Featured
15 min read
#AI#Claude#Data Viz

Latest Writing

Writing Archive

Trump Signal Watcher

social

A live dashboard that watches Truth Social in real time, classifies each post for market-moving signals (tariffs, Iran, China, deals, threats, health), and semantically matches them against a 1,000-market Polymarket political board to surface the trade with a confidence score. Self-improving — a daily agent reviews its own misses and ships fixes.

python polymarket embeddings

Embedded

Open Source

AI-powered voice memo app that transcribes, summarizes, identifies speakers, and makes everything searchable. Native iOS app with a full cloud AI pipeline and open-source connector ecosystem.

swift python openai