Dreeze: pioneering 'Application Revival' for the agentic age

If cloud backup is insurance, recovery is the payout. Yet in the agentic age, the uncertainty surrounding true recovery is only growing. AI agents can now write code and change infrastructure around the clock, making both backup and recovery harder to trust. Recovering data doesn’t automatically mean recovering the application—you can bring back the data while leaving behind the infrastructure, configurations, identities, and connections that make it run.

A cloud-native application is a web of services, accounts, and permissions that changes every day. Traditional recovery was built for a world where restoring the data meant restoring the business. In the cloud, data can come back while the application stays down, because much of what it depends on was never part of the backup. With cloud intrusions up 37% in 2025, that gap is being tested more frequently. Teams are left rebuilding by hand in the middle of a crisis, and crucial hours turn into days. In government systems, the average ransomware attack causes nearly a month of downtime.

This is because agents move faster than any team can track, so last week’s backup may not reflect what’s running today, and last quarter’s recovery plan may no longer work. These agents also bring a new class of risk. They can now act inside production environments, and a single faulty one can break an application without any attacker involved.

In April, an AI coding agent deleted a startup’s production database and its backups in seconds. Startups are cloud-native from day one, lean on agents more heavily, and rarely have a team that can rebuild by hand. As agents take on more of the work, incidents like this will become an operational reality, not an edge case. The security question for every company, whether a fast-moving startup or a global enterprise, is changing from “do we have a backup?” to “can we ensure critical applications will recover?”

Recovery hasn’t kept pace with change, until Dreeze. Bessemer is proud to back the team pioneering “application revival,” recovering the entire cloud-native application and not just the data.

Dreeze’s solution for full application recovery

Dreeze gives companies their applications back when something goes wrong. It starts by mapping the cloud environment and every dependency each application relies on to run. Until recently, building that map meant engineers piecing everything together by hand, and it would become outdated as soon as it was finished. Dreeze leverages AI to do what manual mapping never could: understand how thousands of cloud resources connect, which application each one serves, and how that picture shifts with every change—whether a person or an agent makes it.

This changes what a crisis looks like. Teams don’t have to rebuild the application themselves under pressure; they go back to normal, where everything works the same way it did before the incident. Another problem Dreeze solves has nothing to do with a crisis or a cyberattack. Most teams only find out whether recovery works once they need it. Dreeze has developed “zero-click” testing, where the platform routinely checks that everything is in place, without any dedicated effort from the team. The result is auditable proof of recoverability that companies can use to meet internal SLAs and compliance requirements.

“Once in a shift” opportunity

The opportunity that lies in this shift is enormous. Public cloud spending is expected to exceed $1.4 trillion by 2029, and every critical application running in the cloud needs a reliable way to recover. Data protection is already a multibillion-dollar market, with some of the largest vendors approaching $2 billion in ARR. Yet application-level recovery is still in its infancy. Gartner expects adoption to grow from less than 5% of organizations in 2026 to 35% by 2030. This is the gap Dreeze was built to close.

The team at Dreeze is tackling a critical problem of the agentic age, one that couldn’t be solved properly until now. AI makes the solution possible, and their experience makes it credible. Dreeze is the third backup and recovery company from brothers and co-founders Ziv and Oded Kedem, after building Kashya (acquired by EMC) and Zerto (acquired by HPE). Each company moved recovery forward for its generation of infrastructure, and Dreeze is the next step: recovery built around the application itself.

The Kedem brothers are also joined by former CIO at Zerto and VP of Information Systems at Amdocs, Avi Raichel. Bessemer is co-leading Dreeze’s $33M seed round alongside Norwest to partner with Ziv, Oded, and Avi to ensure that when something breaks, getting it back means it’s really back. Trust doesn’t come easily, and when the backbone of a business is on the line, experience is everything.

If you’re interested in moving forward with Dreeze, join the team.