I'm Sam, a 22-year-old founder who spent the last 8 months building Cardog after a frustrating car buying experience where I felt completely outgunned by information asymmetry.
What it is: An AI platform that gives car buyers and owners the same data transparency that dealers have had for years.
Core features:
AI chat that answers vehicle questions by synthesizing multiple data sources Price analysis that shows if listings are fairly priced based on market comps Digital garage that tracks your car's value and maintenance needs over time
The problem I'm solving: Car buying requires 15+ hours of research across fragmented sources, yet consumers still lack access to comprehensive market data. Dealers know vehicle history, true market values, and maintenance patterns - buyers get marketing copy and hope for the best.
Technical approach: The platform combines structured automotive data (specs, pricing, reliability) with LLM-powered natural language processing. Users ask questions like "Should I buy a 2019 Honda Accord with 60k miles for $23,000?" and get data-backed analysis considering market pricing, maintenance costs, depreciation trends, and model-specific issues.
What I need: I'm looking for a small group of testers (aiming for ~50 people) to try the iOS TestFlight beta and provide honest feedback. The core user flows take about 2 minutes to test:
- Ask the AI a car question to see how it handles complex queries - Upload a vehicle listing (or use our sample ones) to test the price analysis - Add a vehicle to your garage to see the tracking interface
Feedback I'm looking for:
- Does the AI provide genuinely useful answers vs generic responses? - Is the price analysis accurate for vehicles you know well? - What features are missing that would make this valuable for you?
I built this solo while studying CS at University of Toronto. The goal isn't to replace car enthusiast knowledge, but to democratize access to market data and help people make better decisions. TestFlight waitlist: https://cardog.app I'll personally onboard everyone who signs up over the next 48 hours with a quick demo and will be around to answer questions here. A few technical notes for the curious:
- Backend handles real-time data from 15+ automotive sources - Custom ML models for price prediction trained on historical transaction data - iOS app built with Expo / React Native
Thanks for taking a look. Always excited to get feedback from the HN community.