Case Study

Vail, Colorado is best known as a world-class ski destination, drawing millions of visitors each year. But behind the resort economy is a permanent community—teachers, doctors, hospitality staff, and tradespeople who keep the Town running year-round. In one of the most expensive real estate markets in the country, housing that community isn’t incidental to Vail’s success—it’s foundational.

The Town of Vail has spent decades building a deed-restricted housing inventory to ensure local workers can live where they work. These homes limit resale price, occupancy, and/or ownership—creating a separate housing market shielded from second-home and investment demand. Without it, most wage-earning households couldn’t compete. In the free market, even local doctors are priced out. Deed restrictions make it possible for the people who run the community to actually live in it. 

“Each deed restricted home is governed by its own legal agreement,” explains Robyn Smith, Senior Housing Coordinator. “Many written decades apart, with different terms, structures, and levels of clarity.” Historically, interpreting each case has required deep, time-intensive manual review leveraging legacy systems like Laserfiche to ensure rigorous and consistent decisions were made across each case. In many cases, these review cycles took days or weeks to complete.

The stakes are also extremely high. Each deed-restricted home represents a real family's housing stability, and a real community's investment in keeping its workforce intact. Breakdowns in compliance interpretation are not administrative inconveniences. They can determine whether someone stays in their home. "If we get it wrong, people can lose housing stability and the community can lose the capacity those homes were meant to protect," says Dietz.

Addressing this manual administration workload was a problem the Town of Vail had tried to solve before. Over two separate efforts spanning a combined 12 months, staff attempted to manually classify deed restrictions into structured spreadsheet logic. Both failed. The issue wasn't effort, it was the approach. Forcing complex, highly variable legal agreements into rigid categories stripped out the very nuance required to interpret them correctly, creating risk in a process where errors directly affect whether people can stay in their homes.

Kamiwaza is an AI orchestration platform built for organizations that need to work with complex, unstructured information at scale. Unlike traditional approaches that require data to be extracted, cleaned, and restructured before it can be used, Kamiwaza's AI orchestration capabilities bring intelligence directly to the data where it lives, reading and reasoning across documents in their original form. Kamiwaza is also a validated partner in HPE's Unleash AI program, which connects enterprise and public sector customers with proven, production-ready AI solutions built to run in their environment. For the Town of Vail, that combination made all the difference.

The Vail housing team, Kamiwaza, and HPE began by mapping the workflows consuming the most time and resources, identifying deed restriction compliance as a high-impact place to start. Kamiwaza integrated their AI software platform directly with the Town's existing Laserfiche document management system, working inside Vail's environment rather than asking the team to change how they operated. The solution runs entirely behind the Town's firewall on HPE Private Cloud AI infrastructure with NVIDIA RTX 6000 GPUs, with no data migration required and no disruption to existing infrastructure.

A key piece of the solution is Kaizen, Kamiwaza’s premier AI agent. Kaizen reads the full body of each deed restriction document and reasons across it the way a seasoned compliance analyst would. It answers all 38 compliance classification questions with evidence-grounded logic, generates structured compliance reports with zero manual data entry, and gives staff a plain-language chat interface to query the entire portfolio in real time.

Having an AI platform that could quickly read and interpret agreement at scale even challenged some of the team's long-held internal practices. “The biggest challenge wasn’t the technology– it was accepting that we had never been able to unlock this level of consistency and rigor using human interpretation alone,” reflects Jason.

The Town of Vail Housing team has seen on average a 90% reduction in time spent interpreting a deed case, which has had a significant impact on his team. From reducing burnout to unlocking their problem-solving abilities, the housing team is benefiting from AI removing a significant part of the administrative burden. Community members are benefiting as well, receiving answers to their deed inquiries hours instead of weeks. “As leaders,” Dietz reflects,”we should be holding ourselves accountable to giving our teams the tools and insights that return meaningful time back to their day.”

The next phase will extend Kaizen to automatically review annual compliance submittals, eliminating an estimated two to three months of full-time staff work per year. And the roadmap extends well beyond housing. With Laserfiche and OpenGov API connections in place, Vail is building toward a fully integrated agentic platform across municipal operations. "We've only scratched the surface," notes Dietz. "AI gives us a different path. Instead of moving data, we can move intelligence, connecting systems as they are and extracting meaning across them. That allows us to focus on outcomes, not infrastructure."