Ado-Ekiti, Nigeria · open to roles
I sell technical things to people who are right to be suspicious of salespeople. Before this I trained in a laboratory, presented a live youth programme badly and then less badly, and spent eleven weeks teaching myself C++ to build a soil sensor that lied to me for eight of them.
The technical half is not a footnote on the CV. It is the reason I can scope something on the call instead of promising it and apologising later.
Scroll-driven
Everything below grows out of one number: how far you have scrolled. No video, no model file. Seven stages, and six fruit you can actually click at the end.
Your browser can't render the tree, so you're getting the written version. Nothing is missing.
01 — Seed
Rooms I had no particular claim to be in. A cousin's office, a campus society, a producer who needed someone cheap on a Saturday. None of it looked like a career at the time and I would be lying if I said I had a strategy.
Keep scrolling — it grows
02 — Roots
Science Laboratory Technology at Ekiti State University. You do not get to believe a result in a teaching lab, you get to show it twice. That is the whole inheritance, and it is worth more than the certificate.
03 — Foundation
Thirty minutes of youth programming a week. My first episode I read my notes out loud and you can hear the paper. By episode nine I was writing three words per segment instead of sentences, which is when it started to work.
04 — Development
A borrowed Arduino, a voltage divider I had wired wrong, and eleven weeks of a sensor confidently reporting nonsense. I came out of it able to read a datasheet — which has since been worth more in sales meetings than any course I have paid for.
05 — Expansion
We hit the volume target in month four and missed it completely in month five, because I had planned around average rainfall instead of the actual harvest calendar. Plans made in a room fail in a field.
06 — Fruit
Furniture from manufacturing offcuts. Cold chain for tomatoes. If the unit economics need a miracle in cell D14 it goes in the drawer. Three are still alive; one pays for itself.
07 — Impact
They are hanging in the canopy. Click one, or read them further down the page — the fruit and the writing are the same six either way.
Explore my journey
In the order it happened, including the parts that went badly. Pick a stop to ride straight there, or just keep scrolling and let the bus do the work.
The long version
A list of skills is easy to write and impossible to believe. These are the three I would actually talk about in an interview, with the numbers and the mistakes left in.
Cassava aggregation · 2022 — 2023
Nineteen smallholders, one leased truck, and a spreadsheet I was far too proud of. Month four we were eleven percent over target and I started telling people the model worked.
Month five we delivered under half. Not because of rain — because the harvest calendar in three of those villages is set by a rotating labour arrangement I had never asked about. My plan assumed an average. Farming does not do averages.
I spent two weeks doing nothing but sitting with growers and writing down when they actually expected to lift. The rebuilt schedule was uglier, had four separate collection windows, and held for the rest of the season.
2 weeks of asking, to fix a forecast that a month of modelling had got wrong.
Self-taught · 2021 — 2022
I wanted a soil moisture probe that would text me when a bed dried out. Simple project. It took eleven weeks and for eight of them the readings were confident, stable and wrong.
The fault was a voltage divider I had built from a diagram I did not understand. I found it by bisection in the end — halving the circuit, testing, halving again — which is the same thing I had been taught to do with a failed assay, just with a multimeter instead of a pipette.
Nobody paid me for this and nothing shipped. What it bought me is that engineers only have to explain things to me once, and I can tell the difference between a hard problem and an expensive one in a scoping call.
8 of 11 weeks debugging my own misunderstanding, which is the useful part.
“Tell them we're out. I'd rather explain the pipeline gap than the write-off.”
Partnerships · 2023 — 2024
Dinners, two revised term sheets, a site visit. Then I actually read their payment history and found a pattern of stretching suppliers to ninety days and then renegotiating.
I said so in front of the team, which took more nerve than saying it privately and was the right call anyway — a decision made in a corridor is a decision nobody owns. My managing partner did not speak to me properly for about a week.
A competitor signed them four months later and spent eleven months chasing receivables. I do not tell this story because I was clever. I tell it because the four months were real and I want to be the person who wastes them out loud rather than quietly.
11 months of someone else's receivables, avoided by reading one report properly.
Who I am
These are the fruit in the canopy above. Same six, same words — click one on the tree or read them here.
Picks the problem before the solution
Most of what I have added has come from noticing which problem was actually worth solving. The cassava pilot did not need a better spreadsheet, it needed someone to ask nineteen people when they were lifting.
Diagnoses before it guesses
Laboratory training, and a stubborn habit picked up from the Arduino of bisecting a circuit until the fault has nowhere to hide. I isolate the variable, prove the cause, then fix it — a bad deal debugs the same way a bad assay does.
Three words per segment, not sentences
Live television teaches you fast that being right is not the same as being understood. I translate technical work into terms the people funding it or using it can act on, and I cut twelve numbers down to one.
Keeps them warm, and knows when to end one
Four hundred interviews and four years of partnership work: both long games. I keep relationships warm when there is nothing immediate in it, which is exactly why they are there when something comes up — and I know which ones to end.
Good at killing its own ideas
Ten ideas across manufacturing, agriculture and sustainability. Anyone can generate them. The skill is writing the model before the pitch and being honest when cell D14 needs a miracle.
Measured on what shipped
Ideation, planning, execution, coordination, then keeping it running after the launch photograph. A plan that does not ship is a document, and I would rather be judged on the four collection windows that held.
References
A photograph, a clip and a letter you can read in full. All three are placeholders until the real ones are signed off — the names and figures below are fictional.
“She found a rounding error that had been costing us four percent a consignment for two seasons. She was supposed to be learning the CRM that week.”
“By her ninth episode she'd stopped reading and started talking. I've had presenters take three years to do that.”
“She told me to walk away from a distributor I'd spent four months courting. She was right, and she said it in front of the team, which took more nerve than saying it privately.”