Live from the Lab

Labs

Built for real life.

From a live venture arming US retail with AI-driven local search dominance, to a calm companion for new parents and a hiring funnel that re-engages the candidates who've gone cold — these are the products I build, end to end, with Claude Code. Cause-driven, people-first, shipped.

Local marketing · AI search dominance · US retail

GainPlaybook.

Own the search. Own the street.

Live Built with Claude Code
GainPlaybook — Own the search, own the street: the play-route from search to storefront

Sustainable lead generation is engineered — not bought.

The way America finds local businesses changed overnight — AI answers, zero-click results, the map pack. GainPlaybook reads those macro shifts early and turns them into an unfair advantage for US retail.

Macro forces and block-level micro signals, tuned relentlessly through a proprietary, AI-centric system that puts a store where customers actually look: Google, Maps, and the new AI answer engines. Locally, and beyond.

A venture I build and run end to end with Claude Code — the same way this site was.

The four plays

Macro vision, micro precision, AI leverage, compounding pipeline.

01

Read the Macro

The forces rewriting how America searches — mapped before they hit your pipeline.

68% of Google searches now end without a click
02

Master the Micro

Profile, photos, reviews, queries, competitors — tuned block by block.

~48% of local clicks go to the top-3 map pack
03

Dominate AI Search

Show up where customers actually look: Google, Maps, and AI answer engines.

6% → 45% now ask AI where to shop — in one year
04

Compound the Gain

Owned demand that stacks month over month. No rented spikes.

$66.69 — the going rate to rent a single search lead

The field report

The case, in three frames — every number verified against named US-retail sources.

The AI search shift — AI local recommendations grew from 6% to 45% in one year
Search to storefront — discovery is digital, the register is physical
The AI gap is your opening — 45% of consumers ask AI where to shop, most rivals aren't there

Own your local search — built and run end to end with Claude Code.

Visit gainplaybook.com
Parenting · Newborn care

Momma Gets It.

A calm companion for the first, foggy weeks of parenthood.

Live Built with Claude Code

It started the way most real products do — with a problem I was living.

We became first-time parents recently. The early weeks are a blur of feeds, pumps, diapers and half-sleep — and a hundred small questions at hours when no one is awake to answer them.

So I built Momma Gets It: somewhere to log what's happening in seconds, see the patterns when your brain can't, and ask real questions and get answers grounded in actual pediatric sources — never a substitute for a doctor, and always honest about that line.

It was built end to end with Claude Code — the same way this site was. One parent, a clear need, and software that quietly helps.

What it does

Four things, done gently.

Logging a feed in Momma Gets It
Log in seconds

Feeds, pumps, diapers, sleep

One tap from the home screen. Breast or bottle, which side, how long, now or earlier — designed for one hand at 4am.

Lexa answering a question with cited pediatric sources
Ask Lexa

Answers, with their sources

Evidence-grounded help on feeding, sleep and development. Lexa cites UCLA Health, the CDC and the AAP — and is explicit that it is not medical advice.

The Stats dashboard in Momma Gets It
See the patterns

Stats that make sense of the fog

Feeds, volumes, pumped output, sleep, diapers and growth — across 24 hours or 7 days, so trends surface before worry does.

The day-by-day Timeline in Momma Gets It
Every day, in order

A timeline you can hand to anyone

The whole day, one entry at a time — easy to scan, easy to share with a partner, a grandparent, or the pediatrician.

A closer look

Swipe through the app.

Home dashboard
Log a feed
Log a pump session
Log a diaper
Ask Lexa home
A cited Lexa answer
Stats dashboard
Day-by-day timeline

See it in motion

A short walkthrough — tap to play.

Built after becoming first-time parents — entirely with Claude Code.

Visit Momma Gets It
Recruiting · AI prototype · WhatsApp

Nurturing Candidate Engagement.

Re-engage the candidates who go cold.

Prototype Built with Claude Code

Hiring automation only moves people forward. When someone stalls, nothing brings them back.

An assessment left unstarted. A self-schedule link unused. An offer not accepted. The ATS emits no "cold" event — so the funnel quietly bleeds candidates at every stage.

Nurturing Candidate Engagement detects those cold moments and re-engages conversationally on WhatsApp — with recruiters always in control. A 5-minute sweep flags candidates going quiet; a pre-approved template re-opens the conversation; an adaptive AI runs a warm multi-turn chat; outcomes log back to the ATS.

Built end to end with Claude Code, as a working prototype on uberstudio.io.

Nurturing Candidate Engagement landing page — re-engage the candidates who go cold, with The Loop diagram

The Loop

Four moves, from cold signal to recovered candidate.

01

Detect

A 5-minute sweep reads time-in-stage across the whole funnel and flags candidates going cold — computed, since the ATS emits no "cold" event.

02

Reach

A pre-approved WhatsApp template re-opens the conversation. The candidate's reply opens a 24-hour window for free-form, adaptive chat.

03

Converse

Danielle, the AI assistant, runs a warm, personalized, multi-turn conversation — collecting answers or booking time via in-chat WhatsApp Flows.

04

Recover

Logs the outcome back to the ATS and alerts the recruiter. Never advances or rejects — every stage move stays human.

Why it matters

Most candidates don't say no. They just go quiet.

Drop-off is spread across the entire funnel — application, assessment, scheduling, interview, offer. The cost is real, recent, and measurable.

60% start but never finish an application iCIMS, 2025
61% ghosted after an interview Greenhouse, 2024
57% abandon the process mid-way LiveCareer, 2025

Autonomous nudges. Human-controlled decisions.

The agent re-engages on its own but has no tool to move a candidate's stage — advancing, rejecting, or hiring is always a recruiter action. Every send passes a fail-closed compliance gate.

Consent-gated One-tap opt-out 0 agent-authored stage changes Quiet-hours aware US TCPA & EU GDPR

A closer look

The funnel data, the trust boundary, and the sources behind every claim.

Funnel and drop-off stats — candidates going quiet at each stage
Trust boundary — public inputs, synthetic inside, one real external call to Anthropic Claude
Sources — SmartRecruiters API, vendor whitepaper fact-checked, WhatsApp Business platform, ghosting evidence sized from primary sources

What this becomes beyond the prototype

Today the prototype proves the workflow. Production turns the simulated surfaces into live, always-on ones.

Roadmap

Live integrations

Swap the simulated ATS and WhatsApp surfaces for live SmartRecruiters and WhatsApp Business — a binding swap behind the same workflow.

Roadmap

Autonomous scheduling

Move the detection sweep onto always-on cloud crons so re-engagement runs continuously, with no one pressing a button.

Roadmap

Multi-channel reach

Extend beyond WhatsApp to SMS, email, and web chat — meeting each candidate on the channel they actually answer.

Roadmap

Deeper analytics

Layer in recovery-rate, time-to-reconnect, and cohort reporting so the business can see exactly what re-engagement returns.

Built end to end with Claude Code as a working prototype.

Visit uberstudio.io

More on the way

Every build here starts the same way: a real need, a real person, and the conviction that good software should quietly make life easier. New ones land as they're ready.