Cognition introduced Devin Oncall on 15 October 2026, a version of its Devin coding agent that investigates production alerts and bugs, and said in its launch post that it “already handles thousands of alerts every day for some of the world’s largest companies.”
The product page says “Devin answers pages first, escalates to a human when needed, and learns from every alert”. Cognition did not name the companies behind the alert volume or give a figure beyond “thousands”.
Business Pill · AGENT MEMORY
A one-minute explainer of agent memory: how an AI agent keeps what it learned from one task for the next. It teaches the general idea only and says nothing about any person or company in this story.
The key insight: As we read it, Devin Oncall moves Cognition’s coding agent from writing code on request to answering the alerts that wake engineers up, and the one figure Cognition gives is volume: thousands of alerts a day.
Cognition lists three uses on the product page. In alert triage, “Devin follows alerts through infrastructure, logs, traces and code to find the underlying cause”, and when a person is needed it escalates “with the severity, suspected root cause, and recommended next step”.
As an incident teammate, the page says, “Devin joins incident channels automatically and investigates alongside your team”, keeping the timeline, findings and decisions current.
For bug reports and questions, it says Devin reproduces reported bugs and, “When a fix is needed, Devin opens a tested PR with video evidence”.
The tools Cognition’s Devin Oncall product page lists connections to, as of its 15 October 2026 launch: Slack and Teams; the monitoring tools Datadog, New Relic, Sentry, Grafana and Dynatrace (highlighted); and Jira, Linear and ServiceNow.
What Was Posted on X
Introducing Devin Oncall: Devin automatically investigates alerts and bugs, helping your team diagnose and resolve production incidents.
Devin Oncall already handles thousands of alerts every day for some of the worldโs largest companies. Let it become your SRE today. pic.twitter.com/Nza88HwIKd
Cognition’s launch post for Devin Oncall on 15 October 2026, with a product video. It says Devin automatically investigates alerts and bugs and ‘already handles thousands of alerts every day for some of the world’s largest companies’; Cognition does not name those companies in the post.
Memory and Grouping
The page says “Devin connects related signals, groups and deduplicates alerts so the team can focus”, and that it “remembers past failures, to better prevent the same issue from happening next time.”
Its worked example, for an imagined company, shows one investigation covering 14 firings of the same alert, and a memory file noting which team actually fixes an issue when the ownership file in the code repository is out of date.
What Cognition’s devin.ai/oncall product page says Devin Oncall does, as of its 15 October 2026 launch: three listed uses (alert triage, incident teammate, bug reports and questions, in orange) and the behaviours the page describes.
The Tools It Connects To
The product page lists connections to Slack and Teams, to the monitoring tools Datadog, New Relic, Sentry, Grafana and Dynatrace, and to Jira, Linear and ServiceNow.
The page does not give integration counts or usage by tool.
What the Named Customers Say
The page quotes five customers in a section titled “Devin Oncall in the field”. Raunak Chowdhuri, co-founder and CTO of Reducto, is quoted: “For the most part, we don’t have engineers get paged anymore. Engineers no longer complain about being on call.”
Matthew Sweeney, head of support engineering at Modal, is quoted saying the tool “saves me multiple hours of context gathering every day.” Raymond Lin, founding engineer at Crosby, is quoted: “I would rather wait 5 to 10 minutes for Devin to take a first stab at it.”
Clay and AngelList are also quoted, Clay’s Tomi Fiure describing how Devin “reproduces the bug end-to-end, and verifies that its proposed fix actually works.”
What It Costs
The product page gives no Oncall price. Devin’s pricing page lists a free plan, Pro at $20 a month, Max at $200 a month and a Team plan at $80 a month plus $40 a month per full user, and says extra usage on paid plans is “consumed at API pricing.”
The pricing page we read does not say which plans include Oncall.
The Structural Read
As we read it, the product is priced and judged against on-call time rather than coding speed: Cognition’s page leads with ‘Devin answers pages first’, and the Reducto quote it chose is about engineers no longer being paged.
As we read it, the memory feature is the part that compounds. The page says Devin remembers past failures and which team actually fixes an issue, so each alert it handles is meant to make the next one cheaper to investigate.
As we read it, Oncall extends the push we covered on 13 October, when Factory and Cognition shipped rival tools to run agent fleets, from coding work into production operations.
Cognition, Devin Oncall product page (15 October 2026)
“Devin answers pages first, escalates to a human when needed, and learns from every alert”
Three Implications
FOR ENGINEERING MANAGERS Cognition says Devin Oncall already handles thousands of alerts every day for some of the world’s largest companies, without naming them or giving resolution rates.
FOR TOOL VENDORS The product page lists Datadog, New Relic, Sentry, Grafana and Dynatrace among its connections, alongside Slack, Teams, Jira, Linear and ServiceNow.
FOR BUYERS The product page gives no Oncall price; Devin’s pricing page lists plans from free to $200 a month and a Team plan at $80 a month plus $40 per full user.
The framework’s opening line: “SaaS is charged for access to tools. AGaaS charges for the execution of outcomes.”
As we read it, an agent that takes the page, finds the cause and opens a tested fix is sold on the outcome of an incident handled, not on seats. Cognition’s pricing page we read does not say how Oncall itself is billed. The framework does not discuss Cognition or this product.
What Is Not Established
Cognition did not say which companies account for the “thousands of alerts every day”, how many alerts Oncall resolves without a person, or how often its root-cause calls are right, in the material we read. The customer quotes are testimonials Cognition chose to publish.
The worked examples on the page use an imagined company, so they show how the product is meant to work rather than measured results.
Cognition says Devin Oncall takes production alerts first, investigates them through logs, traces and code, and escalates to people when needed, and that it already handles thousands of alerts a day.
Beyond that claim and quotes from five named customers, the launch gives no figures on accuracy, resolution rates or price.
A note on sourcing. We read Cognition’s 15 October 2026 launch post, the Devin Oncall product page in full and Devin’s pricing page. The alert volume and customer names are Cognition’s own claims; the page’s examples use an imagined company. We did not contact Cognition or use other outlets’ reports.
Gennaro Cuofano is a CRO and tech executive who has worked in AI since late 2015, starting with bringing NLP, natural-language generation, voice and chatbot products to the marketing industry. His background is in law and finance: he holds a Master's degree in Law and an International MBA with an emphasis on corporate finance (LUISS Business School and the University of San Diego, 2012), and worked as a financial analyst at a real-estate investment firm in San Diego and as an assistant controller. His work focuses on business model strategy, business engineering and, more broadly, structural analysis: how companies make money, read from their own filings. He created FourWeekMBA and leads research there and at The Business Engineer, his newsletter on AI and business strategy, with over 95,000 subscribers and more than 1,000 published analyses. His writing has also appeared on Entrepreneur, HackerNoon and Search Engine People. Find him on LinkedIn and Substack. See how we source.
Scroll to Top
Discover more from FourWeekMBA
Subscribe now to keep reading and get access to the full archive.
