Founder, High School Firewood · Freshman, University of Georgia
I run a firewood company. I built the software it runs on with AI.
I started High School Firewood at 16 with $500. It is now the #1-rated firewood company in the state of Georgia, and it runs on an operating system I built myself.
Sample orders, not real customers. A simplified version of how my planner loads trucks: it fills each truck by what it can carry, schedules reloads, and won't approve a day when it doesn't know a truck's capacity.
Before this, we used Spoke's free trial. It could route a truck, but it couldn't send one out, bring it back to the barn to reload, and send it out again. This is that day.
8:00 AMSample day
Delivered
0/0
Stacked
0
On the road
0
Customer texts
0 sent
Trucks
Capacity in pieces of wood
Clear a capacity to see the planner refuse the day.
The day
Orders
#
Product
Pcs
Truck
Leaves barn
Arrives
Status
Remove
One order always rides on one truck.
Stop 02 · The barn
One minute at the barn
Stop 03 · Proof of work
The system
A driver app and an operator app, built with AI coding tools, that take an order from the first text to cash in hand. Live on real deliveries since August 31, 2026.
AI front desk
Answers customer texts, books orders into the system, and sends each customer a confirmation.
Clean order intake
Turned 427 raw order submissions into 294 real orders, catching 68 duplicates before a truck left.
Load-aware routing
Plans every route around what each truck can carry. One Saturday it ran 6 loads across 4 trucks, reloads included.
On time
On two recent delivery days, all 14 deliveries arrived inside the window we promised.
Cash custody
Every cash handoff from a driver is tracked and settled, and prices lock the moment a route is approved.
Refuses bad plans
If it doesn't know a truck's capacity, it won't approve the plan. That rule came from a launch that broke.
Stop 04 · The route here
How I built it
Sep 2023
Started the company at 16 with $500, delivering wood from my grandpa's land for gas money.
2024-25
Used ChatGPT to design a four-tier pricing ladder. Orders went from 71 to 229 in one season.
[WINTER YEAR]
An ice storm sent orders in faster than I could route them by hand. Planning one delivery day took about 3 hours.
Summer 2026
Built the driver and operator apps with AI coding tools. My first launch broke because the planner ignored how much a truck could carry, so I rebuilt it around loads.
Aug 31, 2026
Went live on real deliveries. Day planning dropped to about 10 minutes.
Fall 2026
Added AI that answers customer texts, plans every route, and confirms each delivery, so the business keeps moving while I'm at UGA.
Stop 05 · Next route
What's next
The same system for other seasonal service businesses, like holiday lighting, gutter cleaning, and pressure washing, where demand hits all at once and nobody builds software for them.
Shadow mode first
It reads two weeks of an owner's texts, invoices, and schedules, then shows what it would have done and what they would have made, before they pay anything.
Paid from results
I get paid from the extra money it makes the owner, not a flat software fee.
Better every season
Every business it runs teaches it more about pricing and dispatch under a rush.
