# Post Mortem Published articles for Post Mortem. This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles. ## Interning at incident.io: rate limiting, resiliently DevFeed: [Interning at incident.io: rate limiting, resiliently]() Original publisher: [Read original article]() Author: Anthony Oparaocha Published: 2026-09-02T10:33:13Z Content type: article Language: en Sources: [The incident.io Blog]() Topics: [incident](), [systems](), [Cloud](), [dashboards]() Tags: [cloud](), [incident](), [incident-channel](), [incident-management](), [incident-response](), [memory](), [outage](), [post-mortem](), [product](), [production](), [slack-incident](), [systems]() ### AI overview An incident.io intern describes making rate limiting resilient to the loss of its Valkey backing store. The solution used per-pod in-memory top-k buffers so the platform could continue rate limiting instead of failing open when Valkey became unavailable. ### Source excerpt Our rate limiter depends on Valkey. If Valkey goes down we fail open and stop limiting which isn't good enough for our platform. As an intern, I built per-pod in-memory top-k buffers so we keep rate limiting even with the backing store gone. ## We turned off Pub/Sub and nobody noticed DevFeed: [We turned off Pub/Sub and nobody noticed]() Original publisher: [Read original article]() Author: Patrick Hamann; Mike Fisher Published: 2026-08-11T13:56:40Z Content type: article Language: en Sources: [The incident.io Blog]() Topics: [event driven](), [Messaging](), [Publish-subscribe pattern](), [Availability](), [systems](), [Apache Pulsar](), [Cloud](), [Google]() Tags: [availability](), [broker](), [cloud](), [event-driven](), [incident](), [incident-channel](), [incident-management](), [incident-response](), [messaging](), [outage](), [post-mortem](), [production](), [scale](), [slack](), [slack-incident]() ### AI overview This developer article explains how incident.io made its predominantly event-driven platform more resilient by adding a secondary message broker alongside Google Cloud Pub/Sub. It describes the role of message brokers and publish-subscribe processing, the risks of a single point of failure, and the successful production test in which Pub/Sub was turned off without affecting customers. ### Source excerpt Our entire event-driven platform ran through a single message broker, which made it a single point of failure. So we added a second one. This is the story of building an event load balancer, the queuing theory behind it, and the final chaos test where we turned off Pub/Sub in production and nobody noticed. ## Introducing Investigations, powered by Nexus. DevFeed: [Introducing Investigations, powered by Nexus.]() Original publisher: [Read original article]() Author: Pete Hamilton Published: 2026-08-05T13:48:00Z Content type: article Language: en Sources: [The incident.io Blog]() Topics: [incident](), [Incident response](), [Artificial Intelligence](), [telemetry]() Tags: [agentic](), [ai](), [incident](), [incident-channel](), [incident-management](), [incident-response](), [logs](), [metrics](), [outage](), [post-mortem](), [reasoning](), [root-cause-analysis](), [slack-incident]() ### AI overview The article introduces Investigations, an incident.io feature powered by Nexus that autonomously investigates and diagnoses incidents alongside human responders. It analyzes context such as postmortems, logs, metrics, deployments, and dependencies, then provides hypotheses, evidence, next steps, and visible reasoning throughout incident resolution. ### Source excerpt Today we're launching Investigations: agentic root cause analysis that starts the moment you're paged, figures out what broke and why, and works with your team through to resolution. Here's what we built, what's powering it, and why it took some time to get right. ## Institutional knowledge doesn't scale: Building an agentic data analyst DevFeed: [Institutional knowledge doesn't scale: Building an agentic data analyst]() Original publisher: [Read original article]() Author: Navo Das Published: 2026-08-03T10:45:52Z Content type: article Language: en Sources: [The incident.io Blog]() Topics: [data](), [semantic-layer](), [Large Language Model](), [text2sql](), [SQL](), [Databases](), [metric-standardization]() Tags: [agentic](), [analytics](), [building](), [data](), [incident](), [incident-channel](), [incident-management](), [incident-response](), [llm](), [outage](), [post-mortem](), [semantic-layer](), [slack-incident](), [sql](), [tool]() ### AI overview The article explains why dashboard-based self-service analytics and direct LLM access to a data warehouse leave important gaps. It describes building an agentic data analyst, called the "data brain," to distribute institutional data knowledge and help employees ask questions while addressing issues such as canonical joins, filters, metrics, and judgment required to produce correct SQL. ### Source excerpt Institutional knowledge was always the bottleneck. Here's how we built an agentic data analyst to distribute it more efficiently -- and what happened when we let the whole company ask it questions. ## Tailscale didn't stop the Hugging Face intrusion DevFeed: [Tailscale didn't stop the Hugging Face intrusion]() Original publisher: [Read original article]() Author: Avery Pennarun Published: 2026-07-31T18:30:00Z Content type: article Language: en Sources: [Blog on Tailscale]() Topics: [incident](), [AI Bots](), [Vulnerabilities]() Tags: [ai-agent](), [benchmark](), [hugging-face](), [incident](), [kubernetes](), [post-mortem](), [sandbox](), [security]() ### AI overview Tailscale examines the Hugging Face intrusion involving an escaped AI agent, arguing that no Tailscale vulnerability was exploited but that credential controls should have limited the attacker's lateral movement. ### Source excerpt Tailscale wasn't exploited. We still should have stopped the intrusion. ## Behind the Flame: Nicole Hussein DevFeed: [Behind the Flame: Nicole Hussein]() Original publisher: [Read original article]() Author: Megan Batterbury Published: 2026-07-30T14:00:00Z Content type: article Language: en Sources: [The incident.io Blog]() Topics: [incident]() Tags: [collaboration](), [customers](), [engineering](), [incident](), [incident-channel](), [incident-management](), [incident-response](), [on-call](), [outage](), [platform](), [post-mortem](), [product](), [slack-incident](), [technical]() ### AI overview This Behind the Flame profile introduces Nicole Hussein, a Product Engineer at incident.io. It describes her work across product specifications, technical scoping, implementation, customer conversations, and incident-related features, including customizable alert-message templates and ongoing development of the on-call product. ### Source excerpt Meet Nicole Hussein, Product Engineer here at incident.io. 🔥 ## Dashboards aren't (quite) dead DevFeed: [Dashboards aren't (quite) dead]() Original publisher: [Read original article]() Author: Jack Colsey Published: 2026-07-29T16:24:00Z Content type: opinion Language: en Sources: [The incident.io Blog]() Topics: [dashboards](), [Large Language Model](), [semantic-layer](), [metric-standardization](), [data]() Tags: [analytics](), [dashboards](), [data](), [incident](), [incident-channel](), [incident-management](), [incident-response](), [llms](), [outage](), [post-mortem](), [semantic-layer](), [slack-incident]() ### AI overview The article argues that dashboards still matter even as LLMs make flexible, self-serve data analysis increasingly accessible. Semantic layers and reliable interfaces can help humans and LLMs calculate metrics consistently, but dashboards provide a curated, trusted view that keeps the business aligned on which interpretation of the data matters. ### Source excerpt How they still matter as the curated, trusted layer that keeps both humans and LLMs telling the same story from the same data. ## Fact-checking PagerDuty's Opsgenie alternatives comparison table DevFeed: [Fact-checking PagerDuty's Opsgenie alternatives comparison table]() Original publisher: [Read original article]() Author: Tom Wentworth Published: 2026-07-28T14:21:14Z Content type: opinion Language: en Sources: [The incident.io Blog]() Topics: [incident](), [incident management](), [Incident response](), [Slack](), [Artificial Intelligence](), [dashboards](), [Microsoft Teams](), [API](), [Terraform](), [issue tracker]() Tags: [ai](), [api](), [article](), [comparison](), [incident](), [incident-channel](), [incident-management](), [incident-response](), [microsoft-teams](), [outage](), [post-mortem](), [slack](), [slack-incident](), [terraform]() ### AI overview The article fact-checks PagerDuty's comparison table for Opsgenie alternatives, arguing that its description of incident.io is inaccurate. It presents incident.io as an end-to-end AI incident management platform with on-call scheduling, alert routing, incident response, dashboards, mobile access, status pages, investigation, post-mortems, insights, workflow automation, API access, and Terraform support. ### Source excerpt PagerDuty published a new comparison table about incident.io. Once again, it describes a product we don't recognize. So once again, we're correcting the record, row by row, with receipts. ## Read Replica Migration: Lessons and Query Routing Patterns DevFeed: [Read Replica Migration: Lessons and Query Routing Patterns]() Original publisher: [Read original article]() Author: Johanna Larsson Published: 2026-07-21T11:00:45Z Content type: tutorial Language: en Sources: [The incident.io Blog]() Topics: [Databases](), [Database]() Tags: [incident](), [incident-channel](), [incident-management](), [incident-response](), [learnings](), [migrations](), [outage](), [patterns](), [post-mortem](), [read-replica](), [slack-incident]() ### AI overview The article shares incident.io's experience migrating workload to a read replica, including the benefits, operational complexity, and patterns for routing queries between the replica and primary database. ### Source excerpt Our learnings from implementing a product-wide read replica migrations, including some useful patterns for routing queries to replica and primary ## OpenAI and Hugging Face partner to address security incident during model evaluation DevFeed: [OpenAI and Hugging Face partner to address security incident during model evaluation]() Original publisher: [Read original article]() Published: 2026-07-21T07:00:00Z Content type: article Language: en Sources: [OpenAI News]() Topics: [incident](), [hugging face](), [Security](), [Vulnerabilities](), [Artificial Intelligence](), [OpenAI](), [LLM evaluation / benchmarking](), [Incident response](), [Post Mortem]() Tags: [ai](), [evaluation](), [hugging-face](), [incident](), [incident-response](), [openai](), [post-mortem](), [security](), [vulnerabilities]() ### AI overview OpenAI and Hugging Face share findings from a security incident that occurred during AI model evaluation. The updates describe a platform-level compromise, exploitation of a previously unknown Artifactory vulnerability, exposed credentials used by models, and ongoing third-party assessment and incident-response work. ### Source excerpt OpenAI and Hugging Face share early findings from a security incident during AI model evaluation, highlighting advanced cyber capabilities and lessons for defenders. ## Your genie is vanishing: introducing the Opsgenie rescue program DevFeed: [Your genie is vanishing: introducing the Opsgenie rescue program]() Original publisher: [Read original article]() Author: Tom Wentworth Published: 2026-07-09T13:30:00Z Content type: article Language: en Sources: [The incident.io Blog]() Topics: [migration](), [atlassian](), [incident](), [site-reliability-engineering]() Tags: [atlassian](), [incident](), [incident-channel](), [incident-management](), [incident-response](), [migration](), [outage](), [post-mortem](), [slack-incident](), [sre]() ### AI overview incident.io introduces the Opsgenie Rescue Program for customers affected by Atlassian's planned shutdown of Opsgenie. The program offers simplified migration assistance and free overlap, allowing customers to run both systems in parallel without paying two vendors while they validate and complete the transition. ### Source excerpt Today, we're launching the Opsgenie Rescue Program to make that landing soft: simplified migration and free overlap so you never pay two vendors at once. ## Behind the Flame: Ellie Cherrill DevFeed: [Behind the Flame: Ellie Cherrill]() Original publisher: [Read original article]() Author: Megan Batterbury Published: 2026-07-02T14:00:00Z Content type: article Language: en Sources: [The incident.io Blog]() Topics: [coding](), [incident](), [Slack]() Tags: [coding](), [cto](), [customers](), [incident](), [incident-channel](), [incident-management](), [incident-response](), [on-call](), [outage](), [post-mortem](), [slack](), [slack-incident]() ### AI overview This profile follows Ellie Cherrill, a Product Engineer at incident.io, describing her rapid-paced work, which is focused largely on coding and collaboration. She discusses customizing Slack alert messages for customers and contributing to the On-Call system, where changes must be assessed for their impact across the product. ### Source excerpt Meet Ellie Cherrill, Product Engineer here at incident.io. 🔥 ## De-risking a PagerDuty migration: the objections we hear most, and how to clear them DevFeed: [De-risking a PagerDuty migration: the objections we hear most, and how to clear them]() Original publisher: [Read original article]() Author: Eryn Carman Published: 2026-06-09T14:49:00Z Content type: article Language: en Sources: [The incident.io Blog]() Topics: [migration](), [incident](), [Tooling]() Tags: [engineering](), [incident](), [incident-channel](), [incident-management](), [incident-response](), [infrastructure](), [migration](), [on-call](), [organizational](), [outage](), [platforms](), [post-mortem](), [slack-incident](), [switching](), [technical]() ### AI overview This article explains why migrating from PagerDuty to another on-call platform is primarily an organizational and change-management challenge rather than a technical one. It presents seven common objections, including risk of missed alerts, and discusses how to address them through ownership, consensus-building, testing, and parallel operation before cutover. ### Source excerpt Often, switching on-call platforms isn't a technical challenge but a human one. In this post, we break down the seven objections engineering teams raise most often when considering a PagerDuty migration, and share exactly how to address each one. ## Behind the Flame: Pierson Mayhew DevFeed: [Behind the Flame: Pierson Mayhew]() Original publisher: [Read original article]() Author: Megan Batterbury Published: 2026-06-04T14:00:00Z Content type: article Language: en Sources: [The incident.io Blog]() Topics: [incident management](), [Artificial Intelligence](), [SRE](), [Slack]() Tags: [ai](), [emea](), [enterprise](), [incident](), [incident-channel](), [incident-management](), [incident-response](), [outage](), [post-mortem](), [slack](), [slack-incident](), [sre]() ### AI overview This profile introduces Pierson Mayhew, an Enterprise and Strategic Account Executive at incident.io. He describes enterprise sales, collaboration across the company, developing the go-to-market strategy in EMEA, and his excitement about AI SRE and its potential impact on incident management. ### Source excerpt Meet Pierson Mayhew, Enterprise/Strategic Account Executive here at incident.io. 🔥 ## Customers over control: how we measure On-call reliability DevFeed: [Customers over control: how we measure On-call reliability]() Original publisher: [Read original article]() Author: Mike Fisher Published: 2026-05-28T16:29:00Z Content type: article Language: en Sources: [The incident.io Blog]() Topics: [SRE](), [Availability](), [site-reliability-engineering](), [nginx](), [Amazon Web Services](), [API](), [Network](), [incident]() Tags: [2025](), [api](), [availability](), [aws](), [customers](), [http](), [incident](), [incident-channel](), [incident-management](), [incident-response](), [load-balancer](), [metrics](), [network](), [on-call](), [outage](), [post-mortem](), [slack-incident](), [sre]() ### AI overview This article explains how incident.io measures the reliability of its On-call product from the customer's perspective. It focuses on two critical functions, defines SLIs and monthly SLOs, and describes monitoring at the GCP load balancer, alerting, replicated components, and lessons from an AWS outage. ### Source excerpt Instead of thinking about reliability as an exercise in figuring out what we can control, and ignoring anything beyond that, we think about what we'll be really proud to offer to customers. ## Post Mortem: HTTP Request Smuggling Vulnerability DevFeed: [Post Mortem: HTTP Request Smuggling Vulnerability]() Original publisher: [Read original article]() Author: Julien Portalier Published: 2026-05-26T00:00:00Z Content type: article Language: en Sources: [Crystal]() Topics: [Post Mortem](), [vulnerability](), [Crystal](), [HTTP](), [Security](), [Exploit]() Tags: [2026](), [access-control](), [http](), [post-mortem](), [rate-limiting](), [security](), [server](), [vulnerability]() ### AI overview This post-mortem describes an HTTP request smuggling vulnerability in Crystal's HTTP server. The parser mishandled requests containing conflicting framing headers, allowing request injection between a reverse proxy and a Crystal server under specific proxy conditions. The issue was patched in Crystal 1.20.0 and 1.19.2. ### Source excerpt On 12 April 2026, we received a vulnerability report regarding an HTTP request smuggling vulnerability in HTTP::Server. ## Adaptable apps on ChromeOS: a post-mortem DevFeed: [Adaptable apps on ChromeOS: a post-mortem]() Original publisher: [Read original article]() Author: Thomas Künneth Published: 2026-05-23T11:26:02Z Content type: article Language: en Sources: [Thomas Künneth]() Topics: [Post Mortem](), [chromebook](), [Android](), [App](), [Development]() Tags: [android](), [app](), [chromeos](), [coding](), [community](), [dev](), [development](), [engineering](), [inclusive](), [post-mortem](), [software](), [uidesign]() ### AI overview A post-mortem of attempts to remove ChromeOS's resize confirmation for the Be nice Android app in ARC windows. The article explains how the warning appears in windowed laptop mode, why manifest changes did not remove it on Play Store builds, and how tablet mode differs. ### Source excerpt In my previous article Building a custom launcher for ChromeOS I described how Be nice runs on Chromebooks: not as a real default home app, because default home settings (Settings.ACTION_HOME_SETTINGS) usually are not available on ChromeOS, but as a normal Android app in an ARC window. I pretended the app is the launcher in code (detectIsHomeApp() returns true on ChromeOS), worked around split-screen bugs that leave the app a black rectangle, and gave up transparent wallpaper for an opaque scaffold because ARC does not show the ChromeOS desktop behind the window the way a phone shows its wallpaper. What that article obviously could not cover is the fight that came right after. For about a week in May 2026 I tried to get rid of a system confirmation ChromeOS shows when users open the preset-size menu in the ARC window title bar and choose Resizable, at least on Play Store builds. During my experiments, manifest changes often looked fine from Android Studio; however, on Play, the dialog stayed. I tried adding XML, asking AI assistants, including Google's own models, for the magic flag. Here's what I learned along the way. What users see ChromeOS does not present Play Store Android apps the same way in every posture. On a convertible or clamshell Chromebook in laptop use, ARC usually places apps in individual windows (chromeos.dev). The title bar often labels the current layout (Phone, Tablet, or Resizable) and opens a menu to switch presets; choosing Resizable is what triggers the warning below, not dragging the window frame by itself (in fact, resizing the window by dragging one of the window edges does not work until it is allowed through the dialog). In tablet mode the picture is different: apps commonly launch and stay full screen, which matches how ChromeOS has long treated touch-first use on detachables (Chrome Unboxed on immersive mode). The resize presets and the warning below matter most when you are in a windowed, laptop-style layout, not when an app is alre ## Bringing full YAML anchor support to zizmor DevFeed: [Bringing full YAML anchor support to zizmor]() Original publisher: [Read original article]() Author: "Alexis Challande" Published: 2026-05-22T11:00:00Z Content type: article Language: en Sources: [The Trail of Bits Blog](), [The Trail of Bits Blog]() Topics: [GitHub Actions](), [YAML](), [GitHub](), [ci](), [BigQuery](), [bug](), [Open Source](), [dataset]() Tags: [backdoor](), [bigquery](), [bug](), [bugs](), [ci](), [ecosystem-security](), [engineering-practice](), [github](), [github-actions](), [open-source](), [post-mortem](), [pull-requests](), [secrets](), [supply-chain](), [testing]() ### AI overview Trail of Bits describes how it collaborated with zizmor maintainers to add full YAML anchor support to the GitHub Actions static analyzer. The work fixed parsing and deserialization issues, aligned expression evaluation with GitHub's tests, and was validated against 41,253 workflows from 6,612 open-source repositories. ### Source excerpt In March 2026, attackers exploited a pull_request_target misconfiguration in the aquasecurity/trivy-action GitHub Action to exfiltrate organization and repository secrets, then used those credentials to backdoor LiteLLM on PyPI (see Trivy's post-mortem for the full timeline). zizmor is a static analyzer that GitHub Actions users run to catch exactly these misconfigurations before they ship. When GitHub Actions added support for YAML anchors in September 2025, a small but high-value slice of the ecosystem started writing workflows that zizmor could only analyze on a best-effort basis. Over the past three months, Trail of Bits collaborated with the zizmor maintainers to bring zizmor's anchor support up to full coverage. First, we fixed parsing bugs that caused crashes, produced wrong-location findings, and silently mishandled aliased values. Second, we surfaced deserialization edge cases that broke zizmor on otherwise valid workflows. Finally, we helped align zizmor's expression evaluator with GitHub's own Known Answer Tests. We validated all of this against a new corpus of 41,253 workflows from 6,612 high-value open-source repositories. The result: 20 filed issues, 15 merged pull requests. Building the test corpus To understand how anchors are used in CI today and to stress-test zizmor against the full variety of YAML it encounters in the wild, we built a corpus of real workflows. We used BigQuery's GitHub dataset to identify the 10,000 most-starred repositories created between 2022 and 2025, filtered to the 6,612 that use GitHub Actions, and downloaded every workflow file. That gave us 41,253 YAML files. Figure 1: Building a testing corpus When we ran zizmor against the corpus, it crashed on 45 of the 41,253 workflows. That's a low rate, but each crash means a bug in zizmor. How anchors are used in the wild zizmor's anchor support was deliberately limited, and for good reason. YAML anchors make workflows non-local: an alias defined in one place changes behavior else ## Behind the Flame: Maggie Baxter DevFeed: [Behind the Flame: Maggie Baxter]() Original publisher: [Read original article]() Author: Megan Batterbury Published: 2026-05-21T14:00:00Z Content type: article Language: en Sources: [The incident.io Blog]() Topics: [incident management](), [incident]() Tags: [2026](), [company](), [customers](), [incident](), [incident-channel](), [incident-management](), [incident-response](), [outage](), [partner](), [post-mortem](), [slack-incident](), [tech]() ### AI overview Maggie Baxter, a Solutions Engineer at incident.io, describes her work helping technology companies evaluate and implement the company's incident management platform. She discusses relationship-building with prospects and customers, collaboration across the organization, the company-wide offsite, and incident.io's culture of care and personalized customer engagement. ### Source excerpt Meet Maggie Baxter, Solutions Engineer here at incident.io. 🔥 ## Engineering teams in 2027 DevFeed: [Engineering teams in 2027]() Original publisher: [Read original article]() Author: Lawrence Jones Published: 2026-05-19T15:00:00Z Content type: opinion Language: en Sources: [The incident.io Blog]() Topics: [Artificial Intelligence](), [Code](), [Claude Code](), [cursor](), [anthropic]() Tags: [agents](), [ai](), [anthropic](), [claude-code](), [code](), [cursor](), [engineering](), [incident](), [incident-channel](), [incident-management](), [incident-response](), [industry](), [outage](), [post-mortem](), [slack-incident](), [stripe]() ### AI overview The article projects how engineering teams may work with AI by the end of 2027. It describes tailored code agents built around frontier models, fast-starting development environments, integrations with Slack, ticketing and deployment tools, and rich codebase documentation. The author argues that the main investment is in the surrounding environment and integrations, enabling agents to verify changes through compilation, tests and local execution. ### Source excerpt A forward look at where engineering teams are heading with AI, based on conversations with design partners who are visibly six-to-twelve months ahead of the average. Tailored code agents, MCP gateways, agentic products that talk to each other -- most of the picture is already there in pockets, and the rest of the industry is closing the gap fast. ## incident.io launches PagerDuty Rescue Program DevFeed: [incident.io launches PagerDuty Rescue Program]() Original publisher: [Read original article]() Author: Tom Wentworth Published: 2026-05-13T13:30:00Z Content type: release Language: en Sources: [The incident.io Blog]() Topics: [incident](), [Tooling](), [migration](), [Availability](), [Artificial Intelligence]() Tags: [ai](), [availability](), [dependency](), [engineering](), [incident](), [incident-channel](), [incident-management](), [incident-response](), [migration](), [on-call](), [outage](), [post-mortem](), [report](), [slack-incident](), [switching](), [tooling](), [uptime]() ### AI overview incident.io announced the PagerDuty Rescue Program for companies switching their on-call tooling. The program offers contract buyouts, AI-powered migration assistance, built-in migration tooling, a 99.99% uptime guarantee, and AI-first alert investigation. ### Source excerpt incident.io just launched the PagerDuty Rescue Program, making it easier than ever for engineering teams to ditch their decade-old on-call tooling. The program includes a contract buyout (up to a year free), AI-powered white glove migration, a 99.99% uptime SLA, and AI-first on-call that investigates alerts autonomously the moment they fire. ## Humans aren't fast enough for 4 9's DevFeed: [Humans aren't fast enough for 4 9's]() Original publisher: [Read original article]() Author: Norberto Lopes Published: 2026-05-11T17:19:46Z Content type: opinion Language: en Sources: [The incident.io Blog]() Topics: [Availability](), [Resilience](), [incident](), [systems]() Tags: [autonomous](), [availability](), [dependency](), [incident](), [incident-channel](), [incident-management](), [incident-response](), [on-call](), [outage](), [post-mortem](), [recovery](), [resilience](), [slack-incident](), [systems]() ### AI overview The article argues that achieving 99.99% availability is fundamentally different from achieving 99.95%. Because the available recovery window is only 4 minutes and 23 seconds per month, human responders are generally too slow, so systems must provide autonomous recovery, redundancy, and enough resilience to survive critical failures before human intervention is needed. ### Source excerpt Hitting 99.99% isn't a faster version of what you already do. It's a different problem to be solved: autonomous recovery, dependency ceilings, redundancies, and the discipline to build systems that buy you 15-30 minutes before you're needed at all. ## Who's on call? How we improved schedule rendering performance 2,500x DevFeed: [Who's on call? How we improved schedule rendering performance 2,500x]() Original publisher: [Read original article]() Author: Rory Bain Published: 2026-04-28T14:43:46Z Content type: article Language: en Sources: [The incident.io Blog]() Topics: [Claude](), [Algorithms](), [Code](), [Go Language](), [real-time]() Tags: [algorithms](), [claude](), [code](), [cpu](), [go](), [incident](), [incident-channel](), [incident-management](), [incident-response](), [outage](), [post-mortem](), [profiling](), [real-time](), [slack-incident]() ### AI overview The article explains how incident.io improved on-call schedule rendering performance by 2,500x. The system combines historical database entries with future schedule entries calculated in real time, and the improvement came from profiling, algorithmic changes, and Claude Opus 4.6. ### Source excerpt A look at how on-call schedules work, and how we made rendering them 2,500x faster -- through profiling, smarter algorithms, and some Claude. ## How it feels to run an incident with Investigations DevFeed: [How it feels to run an incident with Investigations]() Original publisher: [Read original article]() Author: Chris Evans Published: 2026-04-23T18:06:25Z Content type: article Language: en Sources: [The incident.io Blog]() Topics: [incident](), [User experience (UX)](), [Artificial Intelligence](), [SRE](), [Slack]() Tags: [ai](), [incident](), [incident-channel](), [incident-management](), [incident-response](), [outage](), [post-mortem](), [slack](), [slack-incident](), [sre](), [ux]() ### AI overview This article walks through using incident.io's Investigations, described as an AI SRE and investigation engine, during a real incident. It focuses on improving the incident-response experience through ergonomic UX, automated investigation work, and Slack-based coordination. ### Source excerpt For the last 18 months, we've been building Investigations and one of the things we've learned is that UX matters more than you think. This week, I used AI SRE to run a real incident, and I walk you through it end-to-end. [Next page]()