Uber did not disclose details about its replacement infrastructure.
Just heard that after 12+ years of being a customer, Uber is finally dumping PagerDuty.
PagerDuty used to be the leading paging + incident management tool, went public, then fell asleep at the wheel when Slack came around, then AI. Now everyone tech company I know is moving off.
Many users criticized PagerDuty for poor execution, lost competitiveness, and obsolescence as Uber replaced it after 12 years with a custom AI-built system.
No Digg Deeper questions have been answered for this story yet.
I honestly can’t name a single thing I like about PagerDuty anymore. Their moat got eaten by every adjacent tool adding incident workflows, alerts, uptime checks, and routing.
For smaller teams, it’s hard to justify paying PagerDuty prices when Sentry gives you most of the actual value for way less.
@GergelyOrosz https://incident.io/rescue exists for a reason
They probably literally just vibe coded a replacement and it took a day
Just heard that after 12+ years of being a customer, Uber is finally dumping PagerDuty.
PagerDuty used to be the leading paging + incident management tool, went public, then fell asleep at the wheel when Slack came around, then AI. Now everyone tech company I know is moving off.
They gave us a 75k renewal a couple of weeks back. For 155 users.
I told them to get back to review that because we will vibecode a replacement: We only use it for pages and schedules We keep schedules and users in Terraform We can deploy the coordinator in GCP and AWS to achieve failover. We can integrate with Twilio or whatever
"Codex/Claude build a pagerduty replacement" for a fraction of 75k.
@GergelyOrosz Pagerduty is an example of pretty much every failure mode of non-founder-led companies: - Overly focused on quarterly earnings - No investment in product expansion and no vision for how they could leverage their brand and market position as the market changed
@GergelyOrosz yeah we started using Datadog for paging/incidents lately and it's pretty solid. We already live in Datadog for dashboards/logs/alerts/etc. so it seems pretty obvious to have incident stuff live there too
@GergelyOrosz I can't say who but a major cyber security firm is also leaving PagerDuty for us at FireHydrant, too.
I'm surprised PE hasn't picked them up. But they also have $$$ in debt.
A merger between Elastic and PD makes sense to me at this point.
@GergelyOrosz They didn't let me into their private party at Re:Invent so I guess get rekt nerds
@GergelyOrosz Where are people moving?
@GergelyOrosz We use ZenDuty (which was rebranded a year ago to something I can't remember but was half the price with easier UI)
@GergelyOrosz Rootly ate their lunch
@GergelyOrosz what are folks using these days instead of it?
I will always wonder what PagerDuty got so wrong. Was it going public and being focused on quarterly earnings while rival startups focused on innovation? Did they just get lucky with getting to paging early and never “get” their customers? It feels like a zombie company now
I honestly can’t name a single thing I like about PagerDuty anymore. Their moat got eaten by every adjacent tool adding incident workflows, alerts, uptime checks, and routing.
For smaller teams, it’s hard to justify paying PagerDuty prices when Sentry gives you most of the actual value for way less.
@GergelyOrosz we recently moved off it (moved to datadog) and don’t miss it one bit 🤷♀️.
@GergelyOrosz Did anyone actually use PagerDuty? Seems like @elonmusk was the first to kick them out of Twitter/X. Turns out most of the paging on-call was busy work for bad engineering.
@bitforth @GergelyOrosz I can name one thing, and it's the "servers on fire" choir custom notification alert.
@GergelyOrosz there's probably a subset of customers opting to have hotfixes performed by AI, for better or worse.
@GergelyOrosz What's the best alternative(s)?
@indy_singh_uk @GergelyOrosz Have you actually tried it? http://Incident.io's mobile app sucks and doesn't work nearly as well as PD as a pager. Compare it side-by-side and the difference is obvious.
@GergelyOrosz We simply don't have incidents because our engineers are good.