PagerDuty
Also known as: PagerDuty Advance, PagerDuty Operations Cloud
AI first operations cloud for incident management whose PagerDuty Advance agents (SRE, Scribe, Shift, and Insights) triage, mobilize, resolve, and learn across the full incident lifecycle, with shared agent memory, mandatory human oversight, and a native MCP server across 750 plus integrations.
PagerDuty, based in San Francisco and publicly traded on the New York Stock Exchange, is the operations cloud that serves as the central nervous system for digital operations, and it has become a leader in agentic AI for incident management. Founded in 2009 by Alex Solomon, Andrew Miklas, and Baskar Puvanathasan, former Amazon engineers, and led by chief executive Jennifer Tejada, it went public in 2019 and is trusted by more than thirty thousand companies, including two thirds of the Fortune 100. Its wager is that as software velocity and AI generated code accelerate, the volume and speed of operational data have outrun human led incident response, and the answer is battle tested AI trained on fifteen years and billions of real incidents that does not just analyze but acts.
The platform automates the full incident lifecycle, from detecting and correlating signals across more than seven hundred and fifty integrations to mobilizing the right on call experts, driving resolution, and generating postmortems. Its agentic capabilities live under the PagerDuty Advance brand, built with Amazon Bedrock, Anthropic Claude, and Amazon Q, and now include a suite of agents in general availability: a Site Reliability Engineer agent that triages, correlates alerts, finds related past incidents, and can execute approved runbooks, a Scribe agent that captures incident meetings and drafts status updates, a Shift agent that resolves on call scheduling conflicts, and an Insights or Operations Analyst agent that analyzes historical data to recommend preventative action and automate postmortems. Advance memory shares context across these agents so the system learns cumulatively and works to prevent the roughly sixty percent of incidents that are repeats. Human oversight remains mandatory, with built in guardrails, hallucination checks, and runbook approvals, and a native Model Context Protocol server lets partners and customers connect agents to service, team, and incident context in both directions.
PagerDuty is enterprise grade and open, trusted by Fortune 100 companies to deploy to their compliance, security, and transparency requirements, with an ecosystem of more than thirty AI partners spanning coding agents, observability, and agent governance. It runs as a global cloud operations cloud with regional service options rather than on premise, powers its agents with chosen frontier models rather than a customer picker, and acts through integrations, runbooks, and automation rather than driving a browser. For an operations, site reliability, or DevOps team at scale that wants agentic incident management with deep integrations, MCP native extensibility, and mandatory human oversight, PagerDuty is a strong and proven fit; a small team wanting a simple free alerting tool, or a self hosted platform, may find it more than they need.
Vendor details
Canonical URL
https://www.pagerduty.com
Category
SRE / DevOps agent
Subcategory
AI incident management and agentic operations
Funding status
Independent and publicly traded on the New York Stock Exchange under the ticker PD, headquartered in San Francisco, founded in 2009 by Alex Solomon, Andrew Miklas, and Baskar Puvanathasan, former Amazon engineers, and led by chief executive Jennifer Tejada. PagerDuty went public in 2019, is trusted by more than thirty thousand companies including two thirds of the Fortune 100 and half of the Fortune 500, maintains gross margins above eighty percent, and reports more than three hundred thirty organizations live on its AI agents, with early adopters of its MCP server surpassing two hundred fifty within eight weeks. Customers include Block and Sling TV.
Company status
independent
Use cases & customers
Primary use cases
Target customers
Deployment options
Integrations
PagerDuty connects through more than seven hundred and fifty integrations and a native Model Context Protocol server that works in both directions, letting partners access service, team, and incident context and letting PagerDuty connect to partner MCP servers, alongside direct API integrations. Its AI integrations directory adds more than thirty AI partners across coding agents, observability, IDEs, enterprise copilots, and agent governance, and its agents ingest observability telemetry to act across the stack.
In practice
An anomaly hits at peak traffic and responders lose minutes collecting data and coordinating in a chaotic war room. PagerDuty's SRE agent ingests signals across your integrations, correlates alerts, enriches the incident, and cuts triage from minutes to seconds.
The same failures keep recurring because no one learns from past incidents, and postmortems are a manual chore. PagerDuty's Insights agent analyzes historical data to surface systemic patterns and recommend preventative action, with automated postmortems.
On call coverage gaps and scheduling conflicts create risk and stress before an incident even starts. The Shift agent detects conflicts and finds backup responders automatically, keeping coverage seamless without manual back and forth.
Sources & related URLs
Research sources
Research notes
Added via Crunchbase agentic discovery CSV, enriched full fidelity 2026-07-07. Categorized SRE / DevOps on agentic incident management. Public company (NYSE PD), independent and standalone. Public per user base pricing plus a credit model for AI agents. MCP native leader; batch high alongside Moveworks.
Capability coverage
11.0 / 14 capabilities · 79%
| Integrations & Tool CallingConnects through 750 plus integrations, a native bidirectional MCP server, and direct APIs, with 30 plus AI partners, ingesting observability telemetry and acting across the stack, PagerDuty docs 2026-07-07 | Full |
|---|---|
| Workflow OrchestrationOrchestrates the full incident lifecycle from detect through triage, mobilize, resolve, and postmortem, routing to on call experts across teams, PagerDuty docs 2026-07-07 | Full |
| Knowledge Grounding & RAGTrained on 15 plus years and billions of incidents across 30,000 plus companies, correlating alerts, cross referencing runbooks, and surfacing similar past incidents, PagerDuty docs 2026-07-07 | Full |
| Human Oversight & GuardrailsHuman oversight remains mandatory with built in guardrails, runbooks requiring approval, and agents resolving known issues autonomously while partnering with people on novel ones, PagerDuty docs 2026-07-07 | Full |
| Security, Identity & GovernanceEnterprise grade security trusted by two thirds of the Fortune 100, deployable to a customer's compliance, security, and transparency requirements with built in guardrails, PagerDuty docs 2026-07-07 | Full |
| Observability & AuditabilityIngests observability telemetry for real time incident visibility, analytics, automated postmortems, and a full audit trail, with the Scribe Agent capturing everything, PagerDuty docs 2026-07-07 | Full |
| Memory & State PersistenceAdvance memory shares context across agents and learns cumulatively in the background, drawing on 15 plus years of operational data to prevent recurrences, PagerDuty docs 2026-07-07 | Full |
| Deployment & Data ResidencyRuns as a global cloud Operations Cloud with regional service and residency options, though on premise deployment is not offered, PagerDuty docs 2026-07-07 | Partial |
| Prebuilt Agents, Templates & PacksShips a suite of prebuilt agents including SRE, Scribe, Shift, and Insights, plus an AI use case library and agent templates on the MCP server, PagerDuty docs 2026-07-07 | Full |
| Triggers & Channel CoverageEvent driven core ingests signals and telemetry to trigger incidents on anomalies, with real time multi channel notifications across Slack, mobile, and more, PagerDuty docs 2026-07-07 | Full |
| Model Flexibility & RoutingBuilt on Amazon Bedrock and Anthropic Claude but with no documented customer choice or routing of the agent model, PagerDuty docs 2026-07-07 | Unable to verify |
| APIs, SDKs & MCP ExtensibilityBest in class extensibility with a native bidirectional MCP server, public API, 750 plus integrations, an AI integrations directory, and agent templates, PagerDuty docs 2026-07-07 | Full |
| Testing, Debugging & OptimizationProvides guardrails, hallucination checks, and risk scoring, though a native customer facing agent testing surface is partner provided rather than built in, PagerDuty docs 2026-07-07 | Partial |
| Browser & Computer UseActs through integrations, runbooks, and automation rather than driving a browser or operating a computer interface, PagerDuty docs 2026-07-07 | Unable to verify |
Pricing
Free plan for small teams, then paid incident management from around twenty one dollars per user each month for Professional, with Business and Enterprise above it
per user per month for incident management tiers, with AI agents metered via a credit model
Included quota
Per user incident management tiers, with AIOps and PagerDuty Advance AI agent capabilities as paid additions
What is public
The free plan and per user incident management tiers, starting around twenty one dollars per user each month for Professional, are public, while PagerDuty Advance AI agents use a credit model and Enterprise pricing is quoted.
Billing mechanics
Per user per month subscription for incident management tiers, billed annually, with a free plan at the bottom and PagerDuty Advance AI agent capabilities metered through a credit model layered on top, plus a quoted Enterprise tier.
Cost watchouts
AI agents and AIOps are priced on top of base incident management, often via a credit model that some customers find complex, so agentic usage adds variable cost beyond the per user subscription.
Variable cost rationale
Base incident management is a predictable per user cost, but the AI agent credit model adds usage based variable cost that scales with agentic activity and incident volume.
Additional watchouts
Confirm which tier includes AIOps and PagerDuty Advance, and model the credit consumption of the AI agents at your incident volume, since some customers find the credit model complex.
Overage / add-ons
AI agent and AIOps usage is metered through a credit model layered on top of the per user subscription
Sales call required
Mixed (some tiers require a call)
Free / trial
Free plan for a small team, plus a free trial of paid features
Lowest paid plan
Professional plan around twenty one dollars per user per month, billed annually
Commercial notes
Base incident management pricing is transparent and self serve, but the agentic layer introduces credit based metering that changes the cost model and is worth modeling against incident volume.
Key ambiguities
Base per user tiers are public, but PagerDuty Advance AI agent and AIOps pricing use a credit or consumption model and Enterprise pricing is quoted, so total agentic cost is not fully public.
Cancellation / refund
Monthly and annual billing on public tiers, with a free plan and a free trial of paid features; Enterprise terms are contracted.
Related vendors
- Causely — Causal reasoning engine that infers the single root cause of an…
- Cleric — Safety first AI SRE that investigates alerts read only and delivers…
- incident.io — All in one incident management platform in Slack and Teams with a…
- NeuBird — Autonomous AI SRE built on an Agentic Context Engine that reasons…
- Resolve AI — Most heavily funded autonomous AI SRE, targeting 80 percent…
- Rootly — AI native incident management platform whose AI SRE runs parallel…