Making decisions with incomplete information
Six questions an incident commander can ask to make good calls under pressure, drawn from Apollo 13, the Berlin Airlift, Everest and more.
Rootly Blog
Perspectives on AI, engineering, and the humans keeping systems running.
Incident Management
Your customers will notice before your status page does
Humans of Reliability
Every pilot is ready for engine failure: are your engineers?
Upcoming webinar
Why the Five Whys stop short
Nov 18, 2026 · 2:00 – 3:00 PM EST
In 1912, the Titanic hit an iceberg. Ask why five times and you land on a single root cause, inadequate procedures for ice warnings. Fix that, and you’ve learned almost nothing about why the ship sank, why lifeboats launched half-empty, or why more than 1,500 people died.
Register for the webinar →
Nov 9 – 12, 2026
Rootly in Salt Lake City
Plan your sessions, join our happy hours, and meet the team at booth 340.
Plan your schedule
The unofficial SRE track for KubeCon NA '26
Seven sessions on debugging, testing, and recovering Kubernetes in production.
- When the Clock Lies: How an 8-Minute Time Skew Locked Kubernetes cluster for a day
- GitOps Meltdown: How a Commit Triggered a Cascading Delete of 39 Clusters in Production
- Anatomy of an Outage: When Bursty ML Workloads Break the Kubernetes Scheduler
- Waiting too long to declare
- Asking the reporter to set severity
- Letting the incident commander fix it
- Deciding roles in the middle of the incident
- Chasing root cause before stopping the bleeding
- Saying a guess like it's a fact
- Waiting for the cause before posting on the status page
- Running on-call with too few people
- Asking "why did you do that?"
- Averaging MTTR
- Treating every action item as a promise
- Letting AI make the judgment calls
Incident Management
Why our retrospective template doesn't ask for root cause
The seven sections of the Rootly Retrospective Template, why each one is there, and where AI helps a retrospective and where it should stay out.
Ben On-Call
Is our on-call rotation working, or are we just making it work?
There are 4 of us in our on-call rotation. We’re keeping up with incidents, but taking time off feels like asking everyone else for a favor. How do we know if our rotation is actually working… or if we’re just making it work?
Humans of Reliability
The golden hour: Why the first 15 minutes of an incident decide everything
Twilio incident commander Gandhi Kumar on prioritizing mitigation in the first 15 minutes, customer communication, responder friction, and where AI helps today.
Engineering
Replay, don't rebuild: adding week-long deferrals to a pipeline that handles millions of alerts
On extending a high-volume production pipeline instead of building alongside it, and the edge cases that only appear at the intersection of correct decisions
Engineering
How we used CRDTs to build real-time collaborative retrospectives
A dive into how we built collaborative software with real-time features, AI assistance, and workflow automation with Y.js CRDTs).
Humans of Reliability
Your next internal developer platform is a library of agent skills
PostHog's Michael Kutsch on agent skills as an internal developer platform, team autonomy, AI SRE, and where deterministic guardrails matter.
Incident Management
The titanic didn't sink for just one reason
Why Root Cause Analysis and the Five Whys stop short, told through the Titanic disaster, and what a contributing-factors approach to incidents reveals.
Incident Management
Audit-ready incident records: what SOC 2 auditors look for and how to evaluate tools
How to keep incident records ready for SOC 2 auditors: one record per incident, consistent fields, automatic timestamps and logged changes, plus tool tips.
Incident Management
The hidden cost of vibe coding an incident bot
I built incident bots at two companies, then replaced the one that worked. What vibe coding speeds up, and the ownership costs it doesn't touch.
Incident Management
Datadog vs PagerDuty for incidents (and when to add Rootly)
Datadog On-Call vs PagerDuty compared on paging, schedules, alert sources, status pages, retrospectives and pricing, plus when to add Rootly to either.
Incident Management
Best incident response tools for SRE and platform engineering teams
How platform engineering teams should evaluate incident response tools: adoption, config as code and multi-team governance, with a scenario guide for choosing.
Incident Management
Best AI incident management platforms for 2026
A hype-free comparison of the best AI incident management platforms for 2026—what AI actually ships today and which tools do it best.
AI
Can AI actually find the root cause? Automating postmortems and RCA
What automation reliably contributes to a post-incident review, what it cannot, and how to run retrospectives that produce fixes instead of documents.
AI
Could 2027 be the year you turn off human-in-the-loop remediation?
With reliability platforms as code and agents that rank probable causes from telemetry, catalog and incident history, could 2027 end human-in-the-loop fixes?
AI
Do AI SRE agents actually work? An honest capability assessment
What AI SRE agents reliably do today, what they still get wrong, and how to evaluate a vendor claim without taking the demo at face value.
Product
Why we’re building a channel at Rootly
Rootly’s Partner Program helps MSPs, VARs, consultants, and systems integrators deliver incident response and on-call solutions with flexibility and support.
Incident Management
Real-time telemetry and deploy correlation: how Rootly's AI SRE finds probable root cause in minutes.
How Rootly's AI SRE correlates live metrics, logs and traces with recent commits, deploys and feature-flag changes to post a root-cause hypothesis in minutes.
Humans
Nobody wants to review 1,000 lines of AI-generated code
Engineering leaders from CircleCI, Airbyte, Flare, Moloco, and Twingate reveal what their teams use instead of traditional code review.
Stories couldn’t be loaded.
No stories match these filters. Try another topic or change your search.