Founder Story

A marketer who doesn't code built an elk-call coaching app with Claude Code and says it hit a $650K run rate in 110 days. Every skill learned by sound is the same opening

Ryan Mathews built MtnWispr with Claude Code, with Codex reviewing every change, and paired it with a two-time world champion elk caller's recordings. He says it reached a $650K annual run rate about 110 days after launch (self-reported, not verified). The opening is the same app for every other skill people learn by ear.

Published Nov 11, 2026

The Wow

What someone built

A phone app that scores your elk or duck call out of 100 against the champion's reference calls and gives one plain-English correction, sold as a $99 membership.

Founder Story

Original source
MtnWispr
License
A commercial app; nothing was open-sourced. Build your own scoring app with your own expert and licensed recordings
What you need
  • An expert in the skill who will supply recordings and lend their name
  • A clean set of reference recordings you have rights to
  • An ad budget you can afford to test with
Install
There is nothing to install. Listen to the episode (the build loop is in the first 10 minutes, ads and revenue from about minute 23), try a free scored call on MtnWispr, then pick your own sound-based skill.

Why it's remarkable

The founder says Claude Code alone could not get the scoring right; clean recordings from the expert and a research model's suggestions fixed it. The code was the easy part.

The Opportunity

The business it reveals

Pair with one recognized expert in a sound-based skill, build the scored practice app with coding agents, and split the membership revenue.

Who would pay

Hobbyists in seasonal skills who already pay for gear and lessons, reached through the expert's audience.

The Flip

3 ways you could turn this into a business

Build partner for experts with an audience

Who pays
Coaches, instructors and champions who sell lessons or videos and want an app
How it would earn
A build fee plus a revenue share agreed in writing (our hypothesis)
Why it might work
Experts own the credibility and recordings but not the software; a builder with coding agents can ship a first version fast, as the founder describes.
What already exists
White-label course apps; hiring an app agency.

Free-first-test ad funnel for niche apps

Who pays
Small app founders with a product people can try in seconds
How it would earn
A monthly fee to run creative and ads (our hypothesis)
Why it might work
The founder says sending cold ads straight to a free first test beat the warm-up video; that funnel transfers to other try-it-now apps.
What already exists
Ads agencies; doing it yourself with Meta's automated targeting.

Why we'd try "Scored practice app for another sound-based skill" first: The founder's story shows the moat is the expert and the recordings, not the code, so the next app is a different skill with a different champion, not another elk app.

Build It

How you could build it

A phone-first practice app: pick a call or sound, record it, get a score against the expert's references and one correction, replay the reference, try again, and track progress over time.

First step: Pick one skill people learn by ear, list five recognized experts, and ask one whether they would grade twenty recordings so you can test a score before you build anything else.

Agents and tools

  • Claude Code to build the app and the scoring
  • Codex to review every change
  • A deep-research model to choose audio-scoring methods
  • Hosting such as Vercel or Railway

Setup

  1. Sign the expert and agree the revenue share in writing.
  2. Record clean reference sounds and have the expert grade 20 to 50 practice attempts.
  3. Have the coding agent build scoring until its numbers match the expert's grades.
  4. Ship a free first scored attempt and a paid membership.
  5. Run small ad tests that send people straight to the free attempt, and scale only what pays back.

What the agent does

  • Writes and reviews the app and scoring code
  • Drafts lesson text and ad variations
  • Summarizes member feedback and failed scores

What you do

  • Signs and works with the expert
  • Approves every release and price
  • Runs the ad budget and answers members

Your first customer: The expert's existing students or followers, offered a free scored attempt and a founding-member price.

Test demand first: Score 50 practice attempts and compare them to the expert's grades. Success: the app agrees within ten points on most attempts, and ten people from the expert's audience try the free score in a week.

Your starter instructions

Paste this into the AI agent you use (ChatGPT, Claude, Gemini, Grok Bot or another) as your first message. Copying it starts work in your own agent; it doesn't create or deploy a business by itself.

Get a launch plan

Running it

What stays human: Finding and signing the expert, recording clean reference audio, grading test attempts until the score matches the expert's ear, making the ads and lessons, and supporting members.

Startup cost (estimate)
Under $500 before ads (estimate) (Estimate: AI coding plans plus hosting; the founder calls fixed software costs nominal. The expert's time and recordings are the real cost, usually paid as a revenue share)
Running cost
Mostly ad spend; the founder says he spent a little under $40,000 on ads in September (self-reported) (According to the founder, hosting is a tiny share of revenue and ads are the main cost; your ad budget sets the pace)
How much the agent runs
People run it; AI helps with tasks.

Pricing your own version? Use the free startup-cost worksheet.

The evidence

What the source shows: The founder's own words on the Build With AI podcast (Nov 9, 2026), which we listened to, plus the MtnWispr site and App Store listing. Checked Nov 10, 2026.

What's ours: The other-skill app, the expert build partnership and the funnel service are our applications.

Worth planning around

  • Every revenue figure is self-reported on a podcast; we have not seen a dashboard.
  • The run rate spreads annual plans over 12 months, and revenue is seasonal.
  • Ads drive the business, so the margin depends on ad costs.

Sources

Sources reopened Nov 10, 2026. Spot a mistake? Tell us and we'll correct it in public.

Keep reading

The Daily Paycheck: one remarkable AI business at a time, and exactly how you could build around it. Free; unsubscribe anytime.

Get new AI business discoveries by email.