PagerDuty has introduced Paige, an AI agent for site reliability engineering that is meant to sit in operations, engineering and DevOps teams as one more colleague. It was announced on 3 November 2026. The company's flagship product, until now called PagerDuty Operations Cloud, is renamed the PD Reliability Platform, and Paige runs on top of it. Alongside both, PagerDuty opened PD Labs, an innovation centre built in the open with design partners who get early access to new features.
Paige and the renamed platform are available now. PD Labs begins with three previews. Topology builds a live map of services from code, infrastructure and runbooks without a configuration database. Deep Investigation has Paige test several root-cause theories at the same time and show its reasoning. SRE Workbench is a shared space where people and agents work side by side and where findings are saved to Paige's memory. Of the three, only the Workbench is open for anyone to try today; the other two go to design partners first, who shape them before wider release.
“to fight fire with fire, using AI itself as part of how we keep systems running”
John DiLullo, CEO of PagerDuty · siliconangle.com ↗
How Paige fits into a team is the core of the pitch. It can be placed in an escalation policy as a virtual responder, which means lower-priority incidents may never reach a person. A trigger opens a dedicated channel and conference bridge in Slack or Microsoft Teams, and when the incident is resolved Paige drafts the post-incident review and assigns follow-up tasks with owners and due dates. A new console shows incidents, on-call rotations and AI agents in one view, and only platform customers get it, which gives existing users a reason to move onto the renamed product. Every resolved incident goes into a shared memory that Paige checks when a similar failure appears, and its suggestions arrive with the evidence behind them and a confidence level.
PagerDuty says the reason is the code now reaching production. AI-written software is going live with little human review, and AI-built apps fail in new ways such as model drift, runaway token use, leaked personal data and hallucinations. Chief executive John DiLullo says teams are stretched thin. Autonomy is set team by team: some keep a human in the loop and hold remediation for approval, while others allow Paige to act alone only in well-understood, low-risk cases they have authorised. The company released a benchmark claiming Paige beat leading AI coding agents at diagnosing simulated IT incidents.
On pricing, customers pay a fee tied to their operational scale, with a starting allotment of ingested events, automation actions and seats, and flexible credits for any overage. PagerDuty points to more than $100 million a year in R&D, nearly a billion incidents managed annually, over 750 integrations and the trust of about two-thirds of the Fortune 100. It cites research that 95% of executives see fewer incidents as a competitive advantage, and an IDC finding that IT teams want measurable returns on agent spending such as uptime and time handed back to responders. For a sense of early use, Telus has already built the platform into its internal incident tool, with PagerDuty's Scribe agent running through each incident. Slack's own director of business development, Steve Miller, says the aim is for agentic and human teams to work in the same channel, which is how the Slack and Teams integrations are meant to feel in practice.
Taken together, the launch is PagerDuty's answer to a problem it describes itself: more software is being shipped faster, much of it written by AI, and the people who keep it running have not grown in number. Instead of adding another dashboard, it is offering a teammate that investigates, documents and remembers, while leaving each team to decide how far to trust it. PagerDuty's own research frames the stakes: most executives it surveyed treat fewer incidents as an edge over rivals, and most plan to spend more on resilience in the coming year. Whether customers will let an agent act alone, even in low-risk cases, is something the first months of general availability will show.
About PagerDuty
PagerDuty is an American software company, now based in San Francisco, whose incident-management software helps teams respond to outages. It was founded in 2009 in Toronto by University of Waterloo graduates Alex Solomon, Andrew Miklas and Baskar Puvanathasan, was incubated at Y Combinator, and listed on the New York Stock Exchange in April 2019.
This is not its first AI agent. On 8 October 2025 it launched an agent suite, starting with Scribe, Shift and Insights, which handle call transcription, on-call scheduling conflicts and context-aware recommendations. In May 2026 it named John DiLullo chief executive, with Jennifer Tejada, who had led the company since July 2016, becoming executive chair.
The launch follows earlier steps. In March 2026 PagerDuty said its SRE agent would join escalation policies as a virtual responder, and in October 2025 it released the Scribe, Shift and Insights agents.