About RideVise
Built to help people buy a used car safely — with AI on their side.
No one should have to buy a car alone.
Buying a used car is hard for anyone. It is hardest for people with no one to ask: newcomers to the United States, first-time buyers, people unsure of the mechanics, the paperwork or the language. RideVise exists to put AI on their side, so they can buy safely.
The car-buying market is built for the seller.
When you walk into a dealership, the person across the table has done this thousands of times. They've trained for negotiation. They know which questions to deflect, which numbers to anchor on, and how to make a mediocre deal feel like a steal. Their job is to close. Yours is to not get taken.
RideVise is the AI team on your side. Before you call, before you drive out there, before you sign anything — you'll know what the car is actually worth, what the history shows, which risks are buried in the listing, and exactly how to negotiate based on real data. Not a gut feeling. A full report that reasons over public vehicle records and comparable listings — every finding tied to the source it came from.
We don't take commission from dealers. We don't sell your information to anyone. We don't make money when you buy a car — we make money when our analysis gives you the clarity to buy the right one.
This is what an honest copilot looks like: it gives you the information, you make the call. Every single time.
The information gap is real — and expensive.
I'm Heitor. I came to the United States as a graduate student on an F-1 visa, the way a lot of people arrive here — excited, underprepared for the paperwork, and with no real sense of how the local car market works.
When it was time to buy a car, I realized how alone I was in that process. I didn't know which states had title-washing laws that could leave me holding a car with a fraudulent history. I didn't know that a rebuilt title could void my insurance coverage. I couldn't explain why two nearly identical Civics with similar mileage were $8,000 apart in asking price. When I asked around, every immigrant and newcomer I talked to had exactly the same questions — and the same lack of answers.
The dealership doesn't have that problem. They have a trained sales team, years of transaction data, a legal department, and software that tells them what every car is worth before you walk in. The buyer has Google, Reddit threads from 2018, and hope.
I'm a builder and an AI master's student. I know what these models can do — they can research, reason, and synthesize information across massive datasets in seconds, surfacing insights that would take a human analyst hours to find. So I built the tool I wish I had: something that puts that research power on the buyer's side before they step through that door.
Frontier AI, applied to car buying.
The full RideVise report is being designed as a multi-agent AI pipeline. The plan is for specialized agents to work in parallel: one will compare prices across regional markets and comparable listings, one will turn the vehicle history into plain language, one will prepare negotiation guidance for the specific car and market, one will list the documents your state requires, and one will assemble everything into a clear analysis — with the source of every finding cited.
The pipeline is being designed around frontier AI models — Claude (Anthropic) for deep reasoning and Gemini (Google) for broad data synthesis — chosen for reliability and reasoning depth. It is not meant to be a chatbot wrapper: the goal is a purpose-built system that gives a buyer on their own the kind of insight a professional car buyer would have. The first layer is already live on the home page — a VIN decoder that validates the check digit and reads manufacturer, model year and plant entirely in your browser, with no network call and no language model.
Heitor Alves is a software builder and a master's student in Artificial Intelligence. RideVise applies that engineering discipline to a market where the buyer almost always walks in knowing less than the seller — and it states only what it can show.
When the full report ships, the complexity will stay out of the way on purpose. A buyer shouldn't need to know how many agents ran or which model answered which question — they should walk into the negotiation knowing their number, and exactly why it is the right one.
Beyond the sale.
Buying is just the beginning.
Once you own the car, the information asymmetry doesn't stop. A recall notice arrives in the mail and you're not sure if it applies to your exact trim. The shop tells you the transmission needs $4,200 of work "just to be safe." Your trade-in value has been quietly eroding for six months while you weren't paying attention.
Owner Mode is where RideVise is headed: a companion that stays with you long after the sale. It tracks your car's market value over time and tells you the right moment to sell or trade. It surfaces real recall alerts in plain language and tells you whether your vehicle is actually affected. It helps you evaluate repair quotes against regional market rates before you say yes to anything.
The long-term vision is a full-lifecycle partner — not just a tool for the transaction, but a companion for everything that comes after. Because being on the buyer's side doesn't end when the papers are signed. It's a relationship that lasts as long as you own the car.