First look: behind the scenes building an AI incident responder.
Year in Review 2024
December 17, 2024
2024 was a pretty 🔥 year when it came to incidents, and it's not even over yet!
Whether you were up late fighting fires, handling some tricky comms, coordinating everyone to keep things on track or writing up the post-mortem, tens of thousands of you did it all with incident.io — we're incredibly humbled!
You collectively ran over 100,000 real incidents and spent over 500,000 hours doing it, a staggering 58 years of effort - not bad! Just imagine how much it would have been without us... 😉
We’ll share some more high level insights and our own reflections on the year soon, but in the meantime, we’re delighted to share your own, personalised “Year in Review” - a look back at your incidents in 2024.
We’ve really enjoyed putting it together and we hope you find it interesting — there’s even a sneaky preview of some of the work we’ve been doing around AI to cap it all off at the end 👀
🚀 What else we’ve shipped
New
- When you post a WebEx link in an incident, we'll suggest it as the call for the incident
- Choose whether to sync organisations from Zendesk when configuring the integration
- Schedule overrides are now included in audit logs
Improvements
- Better handling of different email domains when logging in via SAML
- We now have better support for importing multi-layer PagerDuty schedules
- If you're invited to a channel by our product (by being paged or by a workflow), we'll no longer ask you why you're in the channel
Bug fixes
- Using auto-ack during an alert storm was not preventing extra notifications from being sent for auto-acked alerts
- Fixed an issue where expressions in "rich text" links would sometimes fail to parse correctly
- We now skip over Cortex services with invalid groups, rather than failing the whole sync