Ventures
I build what I teach.
I'm not a consultant who has never shipped. The behavioral science and experimentation discipline I bring to Fortune 500 growth teams is the same playbook I use to build and grow my own products.
Building Now
An AI-powered recruitment platform that tailors résumés to each job and automates the application workflow.
35K+ users at $0 current paid user-acquisition spend in July 2026, after earlier paid tests stopped.Visit ↗An experimentation and CRO analytics platform that turns raw A/B test data into behavioral-science insight growth teams can act on.
Built for teams running 30+ experiments a year — self-taught the data pipeline and analytics architecture from scratch to turn scattered test results into searchable institutional knowledge.Visit ↗A fully automated content system built for my own use first — testing whether my voice, writing, and video can turn into genuinely distribution-ready content across social platforms and blogs, instead of productizing it as a SaaS from day one.
In active development — self-taught the automation and LLM-orchestration pipeline solo, and I'm validating it end-to-end on my own content before deciding whether to productionize it further.Practical guides that help analysts, growth teams, and founders make growth decisions they can defend.
Published weekly on Substack.Visit ↗Built & Closed
Not everything I've built is still running. Here's what I shut down — the thesis it started from, what actually happened, and the lesson it left behind. The ones I killed taught me as much as the ones that worked.
My first real company — a test of whether the marketing and economics I was studying actually held up in the real world.
A full e-commerce lifecycle — sourcing, selling, and a 25-person team — wound down when I graduated.Read the teardown →An AI agent that finds the right Reddit conversations and helps you reply with something genuinely useful — turning the best free traffic source into a repeatable channel.
Validated demand and paying customers — then discontinued on principle when it became a cat-and-mouse fight with Reddit's anti-automation.Read the teardown →A privacy-first AI companion built around the real technical problem — keeping an AI's memory of you from drifting or being erased as the models underneath it change.
Built a working prototype and ran deep user research — then chose to sunset it rather than build in a space I couldn't stand behind.Read the teardown →A fast way to make pixel-perfect App Store screenshots — build it in a day, prove distribution in ninety.
A few hundred users and real sales, entirely from Reddit and AI directories — closed as we each moved toward the different halves of the business we most wanted to run.Read the teardown →A dev agency built with a partner to help small businesses implement AI agents and automation — service first, product later.
Built out a partner, a delivery team, and real client-pipeline potential — sunset as we each found our own next direction to build in.Read the teardown →A privacy-first AI chat app, explored in the same wave that produced Myeverly — before checking whether it belonged in a category facing rising regulatory risk with no real differentiation.
Exited before productionizing — the mechanics worked, but the regulatory risk and competitive differentiation didn't hold up once I checked them.Read the teardown →A B2B learning resource for founders and startups on how to grow without burning capital — spend a first funding round on marketing that compounds, not on vanity metrics.
Ran as its own site and newsletter — consolidated once the audience and thesis turned out to overlap almost entirely with what I already wrote elsewhere.Read the teardown →A SaaS concept for cleaning companies — helping short-term-rental hosts vet, schedule, and verify quality across a roster of cleaners — seeded by a good friend's real-world rental-management business.
Stayed in market research — one friend's real pain never closed the gap to a repeatable, paying market.Read the teardown →A publication applying real experimentation rigor — sample size, test duration, design quality — to the behavioral-science literature, to separate what actually replicates from what's just popular.
Sunset as a standalone site — folded into research that's stronger next to an existing audience than standing on its own.Read the teardown →One-click cross-posting from WordPress to Medium — turn every post into syndication, backlinks, and a second audience.
Scoped, built, and shipped inside a single week — earned early signups, then moved on when my co-founder and I read the next move differently, keep pushing distribution or start the next build.An online academy applying research-backed behavioral economics to personal finance — teaching people to out-design their own biases.
Built a 2,900+ community and paid coaching clients, then wound down as I moved on to build the next thing — the behavioral-economics teaching muscle behind everything since.A local SEO consultancy for small businesses — Google, YouTube, and reputation management, delivered by a distributed freelance team.
Ran client work while managing 20+ freelancers with briefs, quality standards, and delivery workflows.Fight online misinformation by putting actual academics on the record — professors interviewed about their own research.
150+ professors contacted, 40+ interviews secured — including Dr. Thomas Simpson of the Federal Reserve Board — but only 10 fully produced; a one-person production and editing workload was the real bottleneck, not the outreach.A portrait photography business built on organic Instagram growth and referral trust — good creative and business fundamentals, wound down once it became clear I wanted to build software and companies more than run a service business.
6K+ Instagram following in year one, a 5-star rating on my Airbnb Experiences offering, and published editorial work — closed when I chose to put that time into building products instead of scaling a one-person service business.How I Build
Every product starts with a behavioral hypothesis, ships lean, and earns its next feature through real experiments — not opinions.
I build solo with AI-augmented development, validate demand before scaling, and treat my own products the way I treat a client's program: name the cognitive mechanism, predict the outcome, measure the lift. It's the fastest way I know to learn what actually moves people.
Ideas I'm Exploring
Beyond what I've shipped, I keep a running set of startup and product ideas — the problems I find worth solving and the first experiments I'd run to test them.
Browse the ideas →Building in growth, experimentation, or AI?
I angel-invest in and advise early-stage teams in these spaces — and I take on a small number of consulting engagements each quarter.
Make Growth Decisions
You Can Defend
Know what to test, when to trust the result, and what to do next. Practical decision guides for analysts, growth teams, and founders.
Opens Substack to confirm · Free · Unsubscribe anytime
Read Better Decisions
Practical guides on experimentation, analytics, commercial judgment, and using AI without outsourcing the decision.
Open Substack (opens in new tab)