Closure Is Building the Search Engine Police Departments Actually Need

The Macro: Law Enforcement Has a Data Problem, Not a Data Shortage

Here is something that surprises people outside of government technology: the average police department is not short on information. They are drowning in it. Body cameras generate hundreds of hours of footage per week. License plate readers capture millions of records. ShotSpotter installations ping audio alerts around the clock. Digital tip lines, social media monitoring, surveillance footage from businesses, cell tower data, financial records. The volume is staggering and it grows every year.

The problem is that none of it talks to each other. A detective working a homicide case might need to cross-reference body camera footage from patrol officers, license plate reader data from three different jurisdictions, cell phone records from a carrier subpoena, and social media posts from multiple platforms. Each of those lives in a different system, often with a different interface, different access permissions, and different search capabilities. Some of it is on paper.

Palantir built a massive business selling data integration to federal agencies and large police departments. Their Gotham platform is powerful and expensive. Axon, which makes Tasers and body cameras, has been building its own data platform. Mark43 and Tyler Technologies sell records management systems. But most of these solutions are designed for large agencies with large budgets. The 18,000 law enforcement agencies in the US include a lot of departments with fewer than 50 officers, limited IT staff, and software budgets that would make a SaaS founder cry.

The clearance rate for violent crime in the US has been declining for decades. In 2023, less than half of violent crimes were solved. For property crime, it was closer to 15 percent. The data exists to solve more cases. The tools to search it effectively do not.

The Micro: Palantir Alum and Israeli Air Force Vet Take on Case Clearance

Closure is building a search and intelligence platform for law enforcement. The pitch is simple: help police departments search across their existing data sources to find connections that solve cases. Instead of a detective spending hours manually pulling records from five different systems, Closure provides a unified search layer.

The founding team has the kind of background that makes this credible. Aaron Zelinger is a co-founder who spent six years at Palantir and two years at Arena. His bio says he “takes helping government seriously; takes himself not so seriously.” That’s a useful disposition for selling to government agencies, which tend to be skeptical of both vendors who don’t understand the mission and vendors who take themselves too seriously. Gilad Levy is the co-founder and CTO. He spent six years in the Israeli Air Force in pilot and cyber security roles, then worked at Spot.io (acquired by NetApp), Microsoft, and Rivery (acquired by Boomi). He has an M.S. in CS and an MBA. That is a deep bench of enterprise infrastructure experience.

They’re a four-person team based in New York, part of YC’s Winter 2025 batch. They’re actively hiring across AI engineering, product design, operations, and data engineering, which suggests they’re in growth mode.

The competitive field is dominated by heavy enterprise players. Palantir is the obvious comparison but operates at a completely different price point and deal complexity. Axon’s platform is tied to their hardware sales. Mark43 and Motorola Solutions sell records management systems that have some search capabilities but aren’t built as intelligence platforms. Cellebrite does digital forensics. ShotSpotter, now SoundThinking, handles audio detection. Closure appears to be going after the search and cross-referencing layer specifically, which is a smart wedge.

The Verdict

I think Closure is attacking one of the most important and most difficult markets in GovTech. Law enforcement technology purchasing is slow, bureaucratic, and relationship-driven. Departments have been burned by vendors who overpromise and underdeliver. The sales cycle can be six to eighteen months. None of this is fun. But if you can actually help detectives close cases faster, the value proposition is impossible to argue with.

The Palantir pedigree is a double-edged sword. It gives the team credibility and domain knowledge, but it also sets expectations. Palantir deals are massive, multi-year contracts with extensive professional services. If Closure is going to serve the mid-market and smaller agencies that Palantir ignores, they need a product that works out of the box without a team of forward-deployed engineers.

At 30 days, I’d want to know how many departments are in pilot programs and what the time-to-value looks like. Can a detective search across data sources within the first week of deployment? At 60 days, the question is whether Closure is generating case leads that actually result in arrests. That is the metric that sells the next contract. At 90 days, I’d watch for department-to-department referrals. In law enforcement, word travels through professional networks. One chief telling another “this thing actually works” is worth more than any sales deck.