GBGene Bealinspillcoffee.hashnode.dev·6h ago · 9 min readOne captured Shopify order drove reproduce, fix, unit test and integration verifyA client’s Shopify storefront sends orders to a dropship manufacturer. Two things were wrong with that integration, and a request to fix one provided me the bandwidth to fix both. The ask was to fix c00
JLJeremy Longshoreinjeremylongshore.hashnode.dev·16h ago · 14 min readEvery Fix Failed in the Shape of the Bug It FixedThe restore drill was right to fail and wrong about why. It was newly added by decision-log/050, and the first snapshot it touched came back bad. It aborted with the loudest explanation it had: the se00
JLJeremy Longshoreinjeremylongshore.hashnode.dev·18h ago · 14 min readEvery Fix Failed in the Shape of the Bug It FixedThe restore drill was right to fail and wrong about why. It was newly added by decision-log/050, and the first snapshot it touched came back bad. It aborted with the loudest explanation it had: the se00
RÇRıfat Çakırinrifatcakir.hashnode.dev·1d ago · 6 min readMocking Spring AI @Tool calls via record/replayThe first question about testing a Spring AI agent is "how do I avoid calling the model." The second one, the moment tools enter the picture, is scarier: When my test makes the agent call refundCusto00
AOAbdullahi Odesanmiinabdspace.hashnode.dev·3d ago · 11 min readTesting a Nuxt Monorepo Without Losing Your MindTesting a standard, single-package app is pretty straightforward. But a Nuxt monorepo? That can get messy very fast. When you're dealing with multiple shared packages, utilities, auto-imports, and all00
RNRaj Nagulapalleinfetchsandbox.hashnode.dev·2d ago · 6 min readGreen passed. The fix granted zero seats.TL;DR "Did the bug stop?" is a weak invariant. An over-suppressing fix — grant nothing — also stops the bug. Assert the exact correct end state: after N identical deliveries, seats == the purchased 00
RÇRıfat Çakırinrifatcakir.hashnode.dev·2d ago · 5 min readYour Spring AI tests are slow, flaky, and cost money. Here's how to make them deterministic.You wire up Spring AI, the ChatClient fluent API feels great, your feature works. Then you sit down to write a test — and every good option is bad. A test that calls a real model is: Slow. A hosted c00
JLJeremy Longshoreinjeremylongshore.hashnode.dev·2d ago · 10 min readThree Commits Between the Rule and the ViolationUnderstanding a failure mode does not inoculate you against shipping it. This blog has returned to vacuous checks repeatedly over the past two weeks, most recently 2026-08-05 and 2026-08-03. One admis00
MVMr. Verifyinkyc-mocking-service-editable-document-templates.hashnode.dev·2d ago · 4 min readArchitecting a Privacy-First KYC Sandbox: Testing OCR Pipelines with Synthetic DataIf you are building a modern FinTech application, a Web3 marketplace, or any platform handling sensitive user onboarding, integrating a Know Your Customer (KYC) API is inevitable. Providers like Onfid00
SVSravan Vidiyalainsravanvidiyala.com·4d ago · 9 min readMy test suite was green. My detector was wrong.Correction, 11 August 2026. The figures below were wrong when first published. My registry harvester paginated over overlapping pages and returned the same servers repeatedly, so I counted 467 servers12K
