New York, United States, October 29th, 2026, FinanceWire
At Riverside, a flawed deploy is detected and reverted in about eight minutes, usually before any customer notices it happened. That turnaround runs on Hud‘s agentic rollouts, a capability Hud has now launched for all of its customers. For Riverside’s engineers, it means less time spent worrying about deployments.
Riverside builds a platform for recording and editing studio-quality audio and video. Its engineers rely on AI to ship quickly, and much of the backend they ship is real-time media code. Code of that kind only reveals its true behavior once it is running under production load, which makes every release an important moment for the user experience.
How Riverside Ships With Hud
Every deploy at Riverside passes through a canary. Layered on top is Hud’s agentic rollout with automatic rollback. When a deploy carries a problem, Hud detects it quickly, and the release is reverted in about eight minutes.
Shilo Magen, VP of R&D at Riverside, described what that means for his team. “Our engineers use AI agents every day, and writing code is no longer the bottleneck. The real challenge is shipping that code to production with confidence.”
He pointed to the nature of Riverside’s product as the reason this matters. “For a real-time media platform like Riverside, even a small regression can have a major impact on the user experience. With Hud, we can automatically detect regressions and roll back problematic releases in minutes. Our engineers now spend less time worrying about deployments and more time building.”
The Launch Behind the Result
Hud’s agentic rollouts allow coding agents to take a change from pull request to production, with Hud validating each step against how the code actually runs. Hud describes an agentic rollout as a rollout built for agents, and it analyzes every change against production before, during, and after the rollout. The aim is to bring the speed of shipping closer to the speed of coding, so safe pull requests reach production as fast as agents write them.
The capability depends on Hud’s runtime code sensor, a one-line SDK that automatically learns the baseline behavior of backend code at the function level. With that baseline in place, Hud works across three moments in a change’s life.
First, before the merge, it reviews the pull request against a production impact map built from real parameters and code flows, and classifies the production risk level. A risky change arrives with clear context that helps reviewers get through it faster, while a safe change can merge as quickly as the agent wrote it.
Second, once the new version is taking traffic, Hud measures its behavior against the baseline. Regressions are caught immediately, and function-level detail reveals the root cause along with the pull request that introduced it.
Third, if a regression occurs, Hud’s context makes it possible to roll back, revert, and fix the underlying issue. The coding agent receives deep context so it does not repeat the error.
For Riverside and other teams, that adds up to knowing the production impact of each change before, during, and after rollout.
A Challenge Shared Across Engineering Teams
Riverside’s experience reflects a wider pattern. Engineers now work with AI agents daily, and those agents can open a pull request within minutes. The slower step comes afterward, when a team needs enough confidence to ship. Pull requests wait for someone willing to merge them, reviewers approve changes they did not have time to read closely, and someone watches dashboards after each deploy. A change written in minutes can wait for days until there is enough confidence to ship it, and the more code agents write, the higher that wall gets. Hud refers to this as the missing link in the AI software factory: agents build fast, but they have been blind to production.
Industry research quantifies the strain. Faros AI’s 2025 vendor survey reported that teams with high AI adoption merge 98% more pull requests while pull request review time rises 91%, meaning engineers who rely heavily on AI produce about twice as many pull requests, each taking about twice as long to review. Harness’s 2025 vendor survey found that only 6% of organizations describe their continuous delivery as fully automated. As agents write more code, the gap between writing and shipping widens, and every team adopting AI runs into it.
Hud’s Perspective
Roee Adler, Co-Founder and CEO of Hud, said the conversation among engineering leaders has moved past code generation. In his view, the merge and rollout are the stages still waiting on people.
“AI already changed how code gets written. The bigger problem now is getting that code into production safely, at the speed agents write it. Every engineering leader I talk to is trying to solve this problem. Coding and triage have agents, but the merge and rollout still run on people, because it’s scary to give coding agents accountability over the stability and reliability of the production environment. That’s the missing link in the AI software factory, and it’s the problem we built Hud to solve.”
He continued: “With agentic rollouts, agents and engineers work from the same runtime context, at the function level, so a change is merged to production based on evidence. When something does regress, it’s immediately detected and fixed, but more importantly the agent that wrote it gets a feedback loop to learn and improve.”
Beyond a Single Customer
Alongside Riverside, engineering teams at monday.com, Lemonade, Cyera, and Drata use Hud, which describes itself as the Runtime Intelligence platform for the AI software factory. Hud’s runtime code sensor learns how backend services behave in production down to the function level, and when a change causes a regression, Hud rolls it back or flags it and gives the coding agent the evidence to fix it.
That feedback loop is where Hud is headed. Each rollout teaches the coding agent how its changes behave in production, setting up the next change to ship with less risk than the previous one.