PinnedShoaib Ahmed Quraishi·Jul 29Why AI Coding Assistants Have Amnesia (And How Knowledge Graphs Fix It)Graphify doesn’t help AI read your code faster — it helps AI stop rereading it.
PinnedShoaib Ahmed Quraishi·Nov 21, 2023Building and Publishing C# NuGet Packages to JFrog Artifactory: A Comprehensive GuideTo create a C# project as a NuGet package and publish it to JFrog Artifactory, you can follow these general steps. I’ll outline the process…
InTowards AIbyShoaib Ahmed Quraishi·9h agoPlaywright Tests Flaky on the-internet.herokuapp.com? Run It Locally in DockerMy failing tests failed again in 10 of 15 reruns against the public site, then passed 60 of 60 locally. Here’s the evidence, the setup, and…
InTowards AIbyShoaib Ahmed Quraishi·2d agoPlaywright MCP vs Playwright CLI vs Test Agents: Which One to Use in 2026Read it for free here.
InTowards AIbyShoaib Ahmed Quraishi·Nov 5Selenium 4.52 Fixed Trust Roots, Not Speed. Should You Still Migrate to Playwright?Selenium 4.52 is mostly trust roots and targeted BiDi improvements, not new testing tricks. Here’s what it fixes, what it cannot, and how…
Shoaib Ahmed Quraishi·Oct 24Accenture Funds Just $29 of Claude a Month. Do You Know What Your AI Spend Actually Ships?Some engineers reportedly burned through $1,000 before the memo landed. A better yardstick, plus a budget guard that holds under retries.A response icon1A response icon1
InTowards AIbyShoaib Ahmed Quraishi·Oct 20Nadella Wants an AI Emergency Brake. Does Your Agent Have One?A fabricated police tip went undiscovered for 72 days. Four controls (stop, record, refuse, confirm what happened) and a hash-chained…
InTowards AIbyShoaib Ahmed Quraishi·Oct 7Your Test Suite Doesn’t Need a Chatbot — It Needs a JudgeI tried to put TypeSafe AI’s Jev through the same evidentiary rigor I’d apply to any system under test. I hit a paywall before I hit a…
InTowards AIbyShoaib Ahmed Quraishi·Sep 8The PR Is Automated. The Review Still Isn’t.Coding agents can produce more PRs than teams can confidently review. The missing discipline is reviewability.A response icon1A response icon1
InStackademicbyShoaib Ahmed Quraishi·Sep 8Nobody’s Testing the AI That’s Testing Your CodeQA agents can now triage failures, heal tests, and skip work. But who measures whether their judgment is actually right?