QQuellix Labs6h ago · 9 min readTest AI Agents Before CRM and Slack Write AccessAn AI agent that says "done" after a CRM update was refused has not merely made a wording mistake. It has given your team a false operational signal. If the same agent can read private fields or act oquellixlabs.hashnode.dev
MMuhammad Ahmad4h ago · 13 min readHow to check an ML benchmark for train/test copiesPowerCons is a small time series dataset: 180 training rows, 180 test rows, two classes. It is one of the 128 datasets in the UCR archive, which researchers use to compare classifiers. 67 of its 180 tmahmadmanzoor.hashnode.dev
PPowerduck4h ago · 5 min readWhen the Client Follows a Redirect and Re-sends Credentials to the Wrong HostQuick answer An HTTP client that follows redirects automatically can carry your credentials to a host you never intended to trust. When a 302 or 307 points at a different domain, a client that blindlyblogs.powerduck.com
PPowerduck4h ago · 5 min readKeeping an Auth Token Fresh Across a Long Test Run vs Re-authenticating Each RequestQuick answer During a long API test run, an access token captured at the start will expire partway through, and a suite that re-authenticates for every single request wastes calls and can hit rate limblogs.powerduck.com
PPowerduck5h ago · 5 min readThe Test Asserted the Entire Response and Broke When the API Added One FieldQuick answer An assertion that compares an entire API response to a fixed literal is testing for exact equality, and it is brittle by design. When the server adds a harmless new field or changes a resblogs.powerduck.com
WWaseem Anjum5h ago · 16 min readVitest vs Jest vs node:test in 2026: Benchmarked on a Real TypeScript ProjectI ran the same TypeScript test suite under Vitest 5.1, Jest 30.5 and Node's built-in node:test, and timed cold runs, warm runs, single files, watch mode and coverage. The suite is 39 test files and 18fabwaseem.hashnode.dev
AAshen Jayawardena11h ago · 26 min readThe Tester's Guide to RAG Evaluation, Part 3: Proving Your Evals WorkIn Part 1 you learned how a RAG system works and fails, then made it reproducible. In Part 2 you built a dataset with a held-out split, found the system's failure modes by reading traces, wrote an evaashenasserts.hashnode.dev
PPowerduck7h ago · 5 min readClient Certificates and mTLS in a GUI Client vs a Spec-Driven FlowQuick answer Mutual TLS lets the server authenticate the client with a certificate instead of a bearer token, and it is notoriously awkward to configure in a GUI client. The certificate and key have tblogs.powerduck.com
PPowerduck7h ago · 5 min readValidating Every Response Against the Contract vs Manual Test ScriptsQuick answer Manual test scripts assert a handful of fields you happened to think of, so they miss the drift that does not touch those fields. Validating every response against the OpenAPI contract chblogs.powerduck.com
PPowerduck7h ago · 6 min readTests with Hardcoded IDs That Collide Against a Shared DatabaseQuick answer Tests that insert rows with hardcoded primary keys, like a fixed customer id of 1001, pass on an empty database and fail the moment they run against a shared or already-seeded one. The fablogs.powerduck.com
