GitHub's Capacity Crisis: Mitchell Hashimoto's Warning to Developers

The green contribution graph of GitHub is starting to show some serious cracks. 📉 If you've felt like your CI/CD pipelines have been lagging recently, you aren't alone. It is pretty apparent how severely GitHub's infrastructure is buckling. The Reality of Uptime: While official status pages report high reliability, third-party monitors show that actual uptime in April 2026 plunged to an abysmal 86%. Late April saw a cascade of severe failures, including the loss of 292 pull requests, a massive botnet attack on the search subsystem, and a critical remote code execution (RCE) vulnerability. The AI Overload: The root cause? The massive rise of autonomous, agentic AI development workflows is essentially hammering the site’s infrastructure beyond its original limits. This instability is actively pushing high-profile maintainers to the exit. When you are deep into configuring advanced, GPU-accelerated terminal environments like Ghostty on macOS, you rely heavily on stable upstream infrastructure. So when its creator, Mitchell Hashimoto, publicly announced he is migrating the project away from GitHub after keeping a journal of chronic outages that blocked his daily work, the community paid attention. 👩💻 If this downtime continues to bottleneck deployments, it is time to seriously look at alternatives like GitLab, the nonprofit Codeberg, or the minimalist SourceHut. #GitHub #SoftwareEngineering #OpenSource #DevOps #WebDevelopment #TechCommunity

GitHub isn’t collapsing, but the shift in developer thinking is very real. This week, Mitchell Hashimoto, creator of Ghostty and HashiCorp co-founder, announced that after nearly two decades on the platform, his major project is moving away from GitHub. The deciding factor? Reliability. Frequent outages, particularly with GitHub Actions, have been disrupting consistent development and deployment workflows. Instead of treating it as occasional inconvenience, he’s taking structured action: Gradual migration to alternatives Keeping a read-only mirror on GitHub Exploring both open-source and commercial platforms This isn’t an isolated reaction. It highlights three growing concerns among engineering teams: • Reliability is now table stakes, not a nice-to-have. When your CI/CD pipeline stalls, velocity drops across the board. • Single-platform dependency carries real risk in 2026. • Flexibility and control are becoming non-negotiable. Many teams are now actively evaluating GitLab, Gitea, Forgejo, and other self-hosted or hybrid solutions. My take as a Full Stack & AI Engineer: Convenience got us here, but production-grade systems demand stability. Whether I’m building distributed backend services with FastAPI + Redis queues or integrating LLM workflows, I can’t afford tooling that blocks progress daily. The future belongs to platforms (and engineers) that prioritize control, observability, and resilience over ecosystem lock-in. Question for the community: If GitHub reliability continues to impact your workflows, would you consider migrating critical projects? What alternatives are you watching? #GitHub #DevOps #SoftwareEngineering #CI_CD #OpenSource #TechTrends #DeveloperExperience

GitHub just had one of the worst weeks in its history. And as engineers, we need to talk about it. Here's what happened 👇 🔴 Incident #1 - The Silent Code Killer On April 23, GitHub's merge queue silently reverted previously merged code across 658 repos and 2,092 PRs - during a 4-hour window. The scariest part? Their automated monitoring caught nothing. They found out via support tickets, 3.5 hours later. The root cause? A change to an unreleased feature that was supposed to be behind a feature flag - but wasn't. The broken code shipped to everyone. 🔴 Incident #2 - The Botnet Blackout On April 27, a suspected botnet overwhelmed GitHub's Elasticsearch cluster. PR lists, issue lists, project views - all blank. For 4+ hours. Data was fine. You just couldn't see any of it. 🔴 Incident #3 - The Uptime Nobody Talks About A developer built an unofficial GitHub status tracker that actually counts degraded performance as downtime (wild concept, right?). Current uptime: 85.51% Industry standard: 99.9% GitHub's official page classifies broken search, PRs not loading, and slowdowns as "Degraded Performance" - technically up, practically unusable. The CTO has now issued a public apology. The reason? Agentic AI workflows pushed GitHub way past its designed limits. They planned for 10x capacity growth. By February, they realized they needed 30x. Three lessons every engineering team should take from this: 1️⃣ Feature flags only work if they're actually enforced - at the infrastructure level, not just in code review. 2️⃣ Monitor for correctness, not just availability. A system can be "up" and completely broken. 3️⃣ How you report incidents is a trust signal. GitHub is now rolling out a 3-tier status system (Degraded / Partial / Major outage) with per-service uptime. That's the right move - just years late. AI-driven workloads are scaling faster than anyone predicted. If it caught GitHub off guard, ask yourself: Is your infrastructure ready? ♻️ Repost if your team uses GitHub. They need to see this. #GitHub #SoftwareEngineering #DevOps #Engineering #IncidentResponse #FeatureFlags #WebDevelopment

GitHub Is Breaking — And Developers Are Done Waiting GitHub hosts 100M+ developers and 420M+ projects. It's not just a tool. It's the public record of software. So when it breaks the whole industry feels it. And right now? It's breaking a lot. Here's what happened in just ONE week: → April 23: 292 pull requests quietly unmerged across 658 repos. Code. Just. Gone. → April 27: A botnet killed GitHub Search for hours. → April 28: The CTO published an apology and a critical RCE vulnerability disclosure on the same morning. Official uptime? They claim 99%+. Third-party monitoring? 86% in April 2026. AWS S3 promises eleven 9s. GitHub is running at one. Then came the moment that changed the conversation. Mitchell Hashimoto creator of Vagrant and Terraform, GitHub user since 2008, daily contributor for 18 years wrote a breakup letter. "I want to ship software, and it doesn't want me to ship software." He kept a journal for a month. Nearly every single day got an X blocked by GitHub outages. His 50,000-star project Ghosty is leaving GitHub. For good. Why is GitHub struggling? Simple: AI Agents are treating GitHub like an unlimited API with no rate limits. Since 2025, agentic workflows exploded and the infrastructure wasn't ready. GitHub isn't just hosting developers anymore. the platform built for developers is now being broken by their AI replacements. Alternatives are ready: → GitLab — reliable, enterprise-grade → Codeberg — German nonprofit, community-driven → SourceHut — minimal, zero AI noise The lesson here isn't just about GitHub. It's about what happens when the infrastructure the world depends on gets treated as a growth asset instead of a responsibility. Reliability is a feature. Not a blog post apology. #GitHub #SoftwareEngineering #DevOps #OpenSource #Developers #AIAgents #TechNews #Microsoft #Coding #CloudInfrastructure #Git #Tech #ProductReliability #ProgrammerLife #MitchellHashimoto

Ghostty just left GitHub. Mitchell Hashimoto's terminal project, gone elsewhere. This is bigger than one project moving its repo. The "GitHub is the default" era of open source is ending. Here's why, and what's actually replacing it. Three forces converging in 2026: 1. AI training rights. GitHub is owned by Microsoft. Microsoft is OpenAI's biggest backer. Every public repo on GitHub is now training data for someone, with terms that aren't 100% clear. Maintainers who care about provenance are voting with their feet. 2. Operational reliability. GitHub had multiple multi-hour outages this month alone. For a tool that gates the deploy pipeline of half the planet, that's not acceptable. Everybody talks about "platform risk" until their CI is down on a Tuesday afternoon. 3. Sovereignty pressure. European companies, government contractors, regulated industries, the "all our code lives on a US-owned platform" model is being questioned. Increasingly answered with: it shouldn't. What's replacing it isn't "another GitHub." It's federated, self-hosted, and pluralistic. Forgejo. Gitea. Codeberg. Sourcehut. Self-hosted GitLab. Even raw Git over SSH for the truly committed. The migration is happening at the maintainer level first, the small-team level next, and the enterprise level after that. The bet I'm making for 2027: open source projects worth caring about will be hosted on something that isn't GitHub. The dependency graph is starting to shift. Pretty sure the next 18 months are going to be loud. #OpenSource #GitHub #Sovereignty #DevTools #Software

Three of the most credible projects in open source either left GitHub or got hit by it in the last 30 days, each for a different reason. At least two of those reasons are not the ones the open source versus GitHub debate has been arguing about. On April 14, Mitchell Hashimoto announced Ghostty is leaving GitHub after 18 years on the platform, pointing at chronic Actions outages and the April 27 incident as the breaking point. His exact line was that GitHub "is no longer a place for serious work if it just blocks you out for hours per day, every day." On April 15, Cal,com pulled its production codebase into a private repository. The community fork on cal,diy stayed open under MIT but lost Teams, Workflows, SSO, Insights, and most enterprise features. Cal,com's stated reason was that AI vulnerability scanners can now systematically pull chains of weaknesses out of open codebases faster than maintainers can patch them. That is AI as an attacker, which is a different argument from the AI training rights story most of the industry has been having. On April 28, GitHub disclosed CVE-2026-3854, a CVSS 8.7 remote code execution vulnerability triggered by a single git push, capable of running code on shared storage nodes that hold millions of public and private repositories. One project left for reliability, one left because of AI assisted attack surface, and the third disclosure is a working example of the kind of vulnerability the second project was trying to get ahead of. Hosting your own code started as a freedom argument, became a price argument, and has now become an argument about not being downstream of someone else's outage or breach. The projects choosing public by default in 2026 are making a different bet than they were making in 2022, and most of them have not noticed the bet changed.

The recent wave of GitHub outages has been more than just a minor inconvenience—it’s becoming a genuine bottleneck for the global dev community. With uptime reportedly dipping below 85% in April 2026 and major disruptions affecting everything from Pull Requests to Actions, the "reliability tax" is getting expensive. +1 When the industry standard for your version control becomes "check the status page before you code," we have a problem. 📉 The Reality Check The "X" Journal: High-profile devs like Mitchell Hashimoto have started tracking daily outages, noting that it’s becoming difficult to treat the platform as a reliable place for "serious work." AI Scaling Pains: GitHub’s move to support 30x capacity for AI-driven workflows (Copilot, agentic AI) seems to be straining the core infrastructure we all rely on. Merge Integrity: Recent bugs in Merge Queues actually caused incorrect commits—a nightmare scenario for anyone managing a production codebase. 💡 Why We Care Git was revolutionary in version control. GitHub took that and put the power of Git on steroids. It was cloud computing before "cloud" was even a buzzword. Over the years, GitHub has added indispensable features like Actions and Issues that have defined the modern CI/CD workflow. 🚀 The Path Forward Microsoft is a large successful organisation and I believe GitHub is in their safe hands despite the recent slip-ups. While there are several hosted Git services like GitLab and Codeberg, the truth is that nothing comes close to our dear GitHub when it’s firing on all cylinders. For CLI paglus like me when it comes to git here is a goodie:https://lnkd.in/guvtzPKq :) Here is hoping that the team can bridge these gaps, stabilize the infrastructure, and put GitHub back on track. We don't want to move; we just want to push code. What’s your take? Have the recent outages changed how you or your team view platform reliability? #GitHub #SoftwareDevelopment #DevOps #Git #TechTrends #CloudComputing

Everyone jokes about rm -rf *… until it actually happens. A while back, GitHub engineer accidentally accidentally ran a destructive command on the wrong repository. Not a fork. Not a personal project. The company’s main GitHub repo. Within seconds… pipelines failed. Services broke. Data disappeared. Panic kicked in. And this wasn’t a small startup. This was at the scale where even minutes of downtime matter. But here’s the part no one talks about 👇 The system came back. Why? Because great engineering isn’t about never making mistakes. It’s about designing systems that survive mistakes. -> GitHub backups saved them -> Branch protections prevented even worse disasters -> Teams jumped in and fixed things fast Within hours, everything was restored. 💡 The lesson? If you’ve ever broken something in code, accidentally deleted a branch, or messed up production… You’re not alone. Even the best engineers have done it. The difference isn’t perfection. The difference is how fast you recover and what you learn. So next time you make a mistake… Don’t panic. Improve your system. Because in tech, mistakes are not the end. They’re part of the process. #github #programming #softwareengineering #devlife #learning #growth #tech

Even the giants have "off" days: Lessons from GitHub’s Merge Queue regression. GitHub recently confirmed a bug where roughly 2,800 pull requests were merged from the wrong base state, unintentionally reverting previous changes. While 0.07% sounds small, in production, "small" percentages can mean major downtime. Key Takeaways for Teams: 1)Automated Testing is King: GitHub is already expanding test coverage for merge operations. 2)Trust, but Verify: Always keep an eye on your branch history after a merge, especially when using automated queues. 3)Transparency Wins: Kudos to Kyle Daigle and the GitHub team for the quick RCA (Root Cause Analysis) and direct outreach to affected users. Have you ever encountered a "silent revert" in your workflow? How does your team guard against tool-level regressions? #GitHub #DevOps #SoftwareEngineering #CI/CD #TechNews

GitHub's CTO Vladimir Fedorov apologized for reliability this morning. Hours later, GitHub disclosed that a single git push could hijack its servers. Two posts. Same morning. Same platform. April 23: a merge queue regression silently reverted 2,092 pull requests across 658 repositories. Squash commits were generated from the wrong base. Merged code appeared to have never existed. April 27: search across PRs and issues fell over, likely under a botnet load. April 28: Wiz researchers showed that a crafted git push could run code on GitHub's servers, outside any sandbox. CVE-2026-3854. CVSS 8.7. Cross-tenant access to millions of repos. github(.com) was patched in 75 minutes. 88% of GitHub Enterprise Server instances are still vulnerable. Git is the boring layer. It is supposed to be deterministic. What you push is what's stored. What's merged stays merged. A push is a push. Three breaches of that contract in five days, on a platform that has not had a CEO in nearly a year. Vlad Fedorov ends his availability post with: "availability first, then capacity, then new features." That order is itself the news. Would appreciate a follow if you want to read more interesting tech stories coming out of this AI era. --- #GitHub #Security #Engineering #Infrastructure #DevOps