Arnas Kazlauskas
Founder & Data Engineer
Why I Built This
As a data engineer, I kept running into the same problem when shopping for a used car: reliability data exists, but it's scattered across government databases, paywalled reports, and anonymous forum posts. The National Highway Traffic Safety Administration publishes detailed recall and complaint records for every vehicle sold in the United States — but their interface wasn't built for comparison shopping.
I built Auto Reliability Index to fix that. It's a system that pulls raw safety data from NHTSA, normalizes it by sales volume, and turns it into simple 0–100 scores you can compare at a glance. No paywalls, no manufacturer sponsorships, no subjective opinions — just data.
The goal is to make vehicle reliability data as accessible and comprehensible as possible, so anyone buying a used car can make an informed decision without needing a data science background.
Background
- Data engineering — building pipelines that collect, process, and score vehicle data from multiple federal and public sources.
- Scoring methodology — designed the four-factor weighted formula (complaint severity, repair costs, recall impact, issue diversity) used across every vehicle on this site.
- Full-stack development — built the entire platform from data ingestion to the consumer-facing interface.
See the data in action