Dextr AI Lands $6.7M to Scale Hotel Agents Handling Bookings, Guest Services and Operations

Sajid Shariff spent years in hospitality before leaving during the pandemic. When he returned to the Bay Area in 2024, he reconnected with industry contacts and discovered the sector faced persistent challenges: labor shortages, rising wage pressures and the constant need to boost revenue.

Shariff began building an AI agent called Alfred for a handful of properties. The system fielded guest inquiries and streamlined check-in and check-out processes. Within three months, both hotels reported gains in their online review scores. As he deployed the agent more widely, Shariff recognized that guest messaging represented just one slice of a much larger opportunity.

This insight led him to partner with Scott Arnold and launch Dextr AI in July 2025 from San Francisco. The startup has now announced $6.7 million in seed funding, led by Elevation Capital with participation from Foundation Capital, marking its first institutional investment. Dextr is building a suite of agents designed to manage reservations, guest communications, staff logistics and other hotel operations.

From guest requests to bookings

Following its initial work in guest management, Dextr developed a voice-based reservations agent named Daisy. Shariff observed that when callers had questions about existing bookings or couldn't locate information on a hotel's website, resolving those questions often converted into new reservations. As hotels integrated Daisy, he found the phone interactions were "leading to revenue."

One large hotel customer now processes between $100,000 and $300,000 monthly in bookings through the voice agent, according to Shariff.

Dextr has since broadened its platform to include agents for group reservations, workforce scheduling and additional functions. The company's vision centers on linking these agents so that data collected in one area of hotel operations can trigger responses elsewhere. For instance, if a guest opts out of housekeeping, the guest management agent could relay this to an operations agent, allowing staff schedules to be modified in real time.

Another application addresses properties without overnight front desk coverage. When a guest arrives late, they could contact the hotel, have Daisy confirm their reservation and get entry instructions. Alfred and Daisy would coordinate to manage the late arrival seamlessly.

For some hotels, automating late check-ins through agents could eliminate the need for an overnight front desk worker. Shariff estimates such a shift costs properties between $70,000 and $80,000 annually in a state like California. "It was very hard to hire for these shifts to begin with," he said.

Building ROI

Shariff identifies three factors that differentiate Dextr from competing hotel AI platforms. While a single agent can execute a specific task, linking multiple agents unlocks broader applications across a property. "It's important that we do multiple use cases to see the higher ROI," he told Crunchbase News.

Second, Dextr's deployment methodology goes beyond simply installing software. Hotels operate differently from one another in how they manage bookings, staff assignments and guest handling. Dextr deploys engineers directly to customer sites to identify where agents can add value and integrate them with existing systems. These engineers remain engaged as agents go live and work to identify use cases with strong return potential. "Every property, every operation is different in this space," Shariff noted. Dextr integrates with property management systems including Oracle Hospitality, OPERA Cloud, StayNtouch, Cloudbeds and Hostaway.

The third distinction involves how staff interact with the agents. Employees can direct them through text or voice using a companion tool called Doss, without needing to master a complex new software interface. High turnover in hospitality makes it difficult for hotels to train staff on intricate systems and retain that knowledge when workers depart. With Doss, "you can call it, you can talk to it, you can text it," Shariff said.

Hotels and more

Dextr's agents now process over 1 million interactions monthly across hundreds of hotels. The customer base spans properties ranging from 21 rooms to 550 rooms, including independent hotels and those operating under Hilton, Wyndham, Best Western and IHG banners, plus reservation centers, vacation rentals and outdoor hospitality venues.

The startup works directly with owners and operators of branded properties and is beginning its first direct engagement with a hotel brand Shariff declined to identify. Dextr also serves hotel management companies, property management firms, reservation centers, vacation rental operators and outdoor hospitality businesses such as campgrounds, RV parks, lodges and golf courses.

Pricing follows a subscription model tied to the property size and agents deployed, with usage-based charges for certain agents. Shariff did not disclose revenue, profitability or the round's valuation. After focusing on pilot programs through much of last year, Dextr accelerated growth starting this spring. Annual recurring revenue has grown approximately 20 to 30 percent month over month, though specific figures were not shared.

The company plans to deploy new capital toward hiring customer-facing engineers, broadening software integrations and creating additional agents. Headcount has expanded from three employees a year ago to 21.

Krishna Mehra, an AI partner at Elevation Capital, said Dextr's early performance attracted the firm's interest. He highlighted customers generating bookings and upsells while cutting operating expenses. "What stood out immediately was Sajid and Scott's execution. Dextr scaled to [hundreds of] contracted properties and meaningful ARR without raising a single dollar," he wrote. "In a space where most AI companies are still searching for product-market fit, Dextr had customers, zero churn, and a clear ROI story. That combination of founder quality and early traction is exactly what we look for at Elevation."