The best products tend to come from people who have been personally annoyed by a problem. Sylvanus Mahe and Kanvaly Fadiga qualify. Before they were the founders of a Y Combinator company, they were two friends running a small car rental fleet, and every week they ran into the same wall: the GPS lived in one app, the fuel receipts in another, the invoices in a spreadsheet, and the answer to a simple question - are we actually making money this month - lived nowhere at all.
Flott HQ is what they built to answer that question. It is an AI-native operating system for transport and logistics fleets: one place that connects the operational side of a fleet (planning, GPS, fuel, mileage, driving hours) with the financial side (revenue, orders, customers, invoices), and then puts AI agents to work on the repetitive tasks in between. The company's own one-line version is plainer: "All your fleet data in one place, with AI agents handling the busywork."
01Where fleet margin goes to die
Ask anyone who runs vehicles for a living and you hear a version of the same complaint. The tools do not talk to each other. A dispatcher plans routes in one system, a driver reports fuel in another, and finance reconciles it all days later, by hand, from receipts and memory. Each tool is fine on its own. The gaps between them are where money quietly leaks - late invoices, idle assets, trips that cost more than they earned, and a picture of the business that is always a few days out of date.
Before — six tabs, one guess
GPS / telematics app Fuel & mileage tracker Dispatch spreadsheet Invoicing software Payments & reconciliation Group chat for the crewAfter — one control center
Planning & dispatch Live tracking Invoicing & payments Collaboration + AI agents Real-time analytics02Five modules, one window
Flott is organized around the daily rhythm of running a fleet rather than around software categories. There is planning and dispatch, where AI helps assign vehicles and drivers and shape the day's missions. There is live tracking of position, fuel, mileage, and hours. There is invoicing and payments, with dynamic pricing and follow-through on who has paid. There is a collaboration layer where the AI agents live, and analytics stitching the operational and financial numbers into one view.
Planning & Dispatch
AI-assisted route planning that assigns vehicles and drivers and optimizes each mission.
Live Tracking
Real-time position, fuel consumption, mileage, and driving hours across the fleet.
Invoicing & Payments
Automated billing with dynamic pricing and payment tracking to shorten the cash cycle.
Collaboration & AI
Team coordination where AI agents take on repetitive operational and financial tasks.
The unlock is the fifth piece - analytics that read from both halves at once. Most fleet software answers "where is the truck." Flott is built to answer "where is the truck, and is this trip making us money," which is a harder question and a more useful one.
The customers03Anyone who runs assets that move
Flott is not built for one narrow niche. The same problem - scattered tools, blurry margins - shows up wherever a business runs a fleet. That includes car rental companies and taxi and ride-hail operators, trucking and last-mile logistics, and the more industrial end of the map: construction, waste collection, and mining. What they share is not the cargo but the shape of the pain.
04Growth without a marketing budget
At launch, Flott reported roughly $200,000 in annual recurring revenue, growing around 60% month over month - and all of it, by the founders' account, came through word of mouth rather than paid marketing. For an early company that number matters less as a milestone than as a signal: when operators refer a tool to other operators before anyone asks them to, it usually means the tool solved something real.
05High school friends who never stopped building
Sylvanus Mahe (co-founder and CEO) and Kanvaly Fadiga (co-founder and CTO) have been building things together since they were teenagers. Both are engineers with operations experience, with backgrounds that pass through McKinsey and Datadog - strategy on one side, data at scale on the other. Then came the car rental business in Cote d'Ivoire, which taught them the part no consulting deck covers: what it actually feels like to dispatch your own vehicles at dawn and chase your own invoices at night.
That combination - lived operational pain plus the engineering to fix it - is the company's real asset. They are not building fleet software from the outside in. They are building it from the garage out.
The market06A new angle on an old category
Fleet management is not an empty field. Incumbents like Samsara, Motive, Geotab, Fleetio, and Verizon Connect have spent years wiring up telematics and compliance. Flott's wager is not that those tools are bad, but that they mostly solve one slice - tracking, or dispatch, or logs - and leave the operator to bolt the finances on afterward. By starting AI-native and putting operational and financial data in the same system, Flott is aiming at the seam the incumbents left open.
Traditional fleet stack
Tracking OR dispatch Finances handled elsewhere Reports after the fact Humans do the busyworkFlott's approach
Ops + finance in one system AI-native from the start Real-time, unified view Agents do the busyworkWhether Flott becomes the default operating system for fleets or a sharp tool for a particular kind of operator is the open question every early company faces. What is already clear is the shape of the idea: take an industry that runs on disconnected software and one person's Sunday spreadsheet, and hand it a single window where the trucks and the money finally line up.