Vlad also worked on the Google Maps API as well as ridesharing and travel. Yarik led ML/Search infrastructure at Apple, Google, and Meta powering products that are used by millions of users every day. We realized no place was treating data freshness as an infrastructure, so we’re building it.
We started with one of the hardest parts – knowing if a place is even real. Our Business Validation API (https://github.com/voygr-tech/dev-tools) tells you if a business is actually operating, closed, rebranded, or derecognized. We aggregate multiple data sources, detect conflicting signals and return a structured decision. Think of it as continuous integration, but for the physical world.
Problem: ~40% of Google searches and up to 20% of LLM signals include local context. 25-30% of places churn every year. The world doesn’t emit structured “I turned off” events – you have to actively seek it out. As agents start searching, booking, and buying in the real world, this problem becomes 10 times bigger — and no one is building the infrastructure for it. We recently benchmarked how well LLM handles local location queries (https://news.ycombinator.com/item?id=47366423) – the results were poor: even the best 12 local location queries got 1 wrong
We are processing thousands of locations per day for enterprise customers including major mapping and technology companies. Today we’re opening up API access to the developer community. Please find details here: https://github.com/voygr-tech/dev-tools
We’d love honest feedback – whether it’s about the issue, our approach, or where you think we’re wrong. If you’re dealing with old location data in your own products, we’d especially love to hear what breaks. We’re here all day, AMA.
<a href