It's not easy to turn terabytes of raw datafiles into a performant and queryable index of records. How do we know if two records from two data sources refer to the same entity? What should the merge strategy be if they are? How do we handle aliases or typos (e.g. B., Bil, Bill, Billy, William or St., Street). What if address components are out of order, or missing? Where and how do we even amass high-quality data in the first place?! These are the kinds of problems we work on solving every day. Durt Holdings is a data technology company founded in 2016 with the goal of aggregating and productizing publicly available data. Our core technology, the Durt Data API, does the heavy lifting of ingesting billions of public records to serve as the foundation for all of our consumer facing products. When we make a new dataset available, we build a small team dedicated to making a consumer product around it. Any kind of data, we are digging for it and building on it. Suppose you want to know if your date has a criminal history. Premium and expensive background checks are where most people start but nobody wants to pay $50 every time they go on a date. Sometimes its also difficult to verify the person you are meeting is actually who you requested the background check for; there are a lot of John Smiths in Washington and often the results won't have pictures. Seems expensive for a "maybe", even for a "probably". Durt's solution is instead to source criminal records directly from public government-maintained registries and make it available for free in one easy to use search engine. The only problem is, there are thousands of sherrif offices and hundreds of prisons both state and federal. It's all public data, its just hard to aggregate. That's our business. Same deal for persons, vehicles, homes and more.