Başar BallıözProduct Engineer
← All posts

Why I built RPMBase

Every car owner carries two questions around: what should I do next, and what has this car really cost me? RPMBase is built to answer both.

By Başar Ballıöz · · 3 min read

How it started

Cars are expensive in small pieces. Fuel one week, parking the next, then tyres, insurance, an inspection, an oil change. Because the costs arrive one at a time, the total never feels real. The first time I sat down and added up a year of receipts, the number was clearly higher than the one in my head.

The other half of the problem is timing. Most cars come with a service book that nobody keeps up to date, so the next thing that is due tends to be discovered when it is already late: an inspection date that slipped by, an insurance renewal noticed at the last minute.

I had just built RPMVault for motorcycles, and the question I kept asking myself was why drivers did not have the same thing. They did not need a catalog. They needed their own car's history in one place and a clear answer to "what now?"

Why a new app and not a bigger RPMVault

Adding cars to RPMVault would have been the quick route. I decided against it. A rider researching bikes and a driver keeping up with a family car want different things from the first screen, and one product trying to serve both would have served neither well.

So RPMBase starts from the garage, not the catalog. Its main screen answers one question: what should I do next?

What I built

  • The car at a glance. Plate and VIN, insurance and inspection dates, mileage and spending, with quick actions to log fuel or a service.
  • Activity with countdowns. Oil changes, insurance, inspection and documents, each with how long is left, plus one-tap logging.
  • Costs that add up. Spending split into service, fuel, parking and more, so the cost of owning the car becomes a real number.
  • The car's whole story. Every milestone becomes a timeline you can share or export as a PDF, which helps when it is time to sell.
  • An assistant grounded in your records. It answers from the car's own history, and it reads receipt photos into entries so you do not have to type them.

The decisions that shaped it

Typing is the enemy

A log only works if people keep it, and people stop keeping logs that feel like homework. Every entry should take seconds. That is why receipt scanning and quick actions mattered more to me than another chart.

Four languages from the start

The app and the Next.js website are in English, Turkish, Spanish and German. The site has guides and calculators for everyday maintenance questions, which is also how many people find the app.

A plan that feels fair

RPMBase is free for one car. Premium covers up to five cars and a higher monthly allowance for the assistant. Someone with a single car should be able to use it fully without paying.

Reusing what worked

React Native and Expo for the app, and a Node.js API with Firebase authentication behind it. Much of what I learned on RPMVault, from data models to store listings, carried straight over. That is a big part of why a one-person project can move faster the second time.

What I learned

Reminders beat reports. Charts are nice to look at once. What keeps a car in good shape is a quiet reminder at the right time, so the countdowns got more attention than any graph.

An assistant has to know its limits. Grounding answers in the car's own records makes them useful; being clear that it is a helper and not a mechanic makes them safe. For anything about brakes, tyres or a warning light, the right answer is still a qualified workshop.

The second product is easier to build and harder to explain. Everyone already knows what a car app is. The real work is showing, in a few seconds, why this one is calmer and clearer than the notebook in the glovebox.

After two apps about vehicles, I wanted to make something completely different: a game, for children. That is Tibblo.

More posts

Godot on iOS: from an uneven 60 fps to 120 Hz with MetalA Godot 4 game stuck near 60 fps on an iPhone 14 Pro with a main thread busy only 4 ms a frame. How a phone bench and Instruments found the cause, and how switching iOS to Metal gave 119 fps. Quest pacing: measuring where a 150-level game goes flatHow I measured the story pacing of Tibblo's 763-day road from level 1 to 150, found where it went flat after level 40, and fixed it with three-step arcs, a daily limit and a 'Last time…' card.