Ttappers·1d agoFail-Before-Fix in AI Coding Agents: Speculative Patching vs. a Reproduction-First Gate (-95.3% Input Tokens, 100% Accuracy)💡 AI Labo Benchmark Series — Experiment E22 (Verification Gates & Bugfix Workflows) This post is part of our empirical benchmark series testing AI coding agent harness patterns inside network-isolate9 min read·tappers.hashnode.dev
Ttappers·4d agoWhy AI Agents Write Weak Tests for Their Own Code: Single-Agent TDD vs. Two-Agent Blind TDD (0% vs. 100% Mutant Kill Rate)💡 AI Labo Benchmark Series — Experiment E06 (Verification & Mutation Testing) This post is part of our empirical benchmark series testing AI coding agent harness patterns inside network-isolated Prox8 min read·tappers.hashnode.dev
JJulian Tellez·Oct 23Stop Letting AI Mark Its Own HomeworkLetting an AI review its own code is marking its own homework. How Greenlight enforces a TDD boundary between test and code agents.7 min read·juliantellez.hashnode.dev
RRandi W·Oct 18Heroics are not a deploy planOn rubber-stamp approvals, test classes written at midnight, estimates nobody practiced, and what changed when we stopped relying on heroes.9 min read·salesforcegirl.hashnode.dev4AA
PPablo Curell Mompo·Aug 19The wrong unitShipping sooner, not faster10 min read·blog.codemanship.dev22K
RRajshree·Aug 20The AI-Augmented Developer: A Practical Guide to Refactoring and TDD with CursorLearn how to integrate AI tools like Cursor and Claude Code into your daily software engineering workflow. A practical tutorial on refactoring legacy code and AI-driven TDD.14 min read·rjshree.hashnode.dev
JJeremy Longshore·Jul 1FTS5 Fallback: How Zero Search Results Became Five (ICO Dogfood Day One)Five questions in. Zero answers out. The corpus had every answer; the search layer refused to admit it. Twenty-four hours later, against the same compiled workspace and the same five questions, ICO re10 min read·jeremylongshore.hashnode.dev
SSho Naka·Jun 16Tests Pass, Design Breaks: Why TDD Can't Hold the Line on Design IntentTDD verifies the implementation matches the tests. It doesn't verify the tests match the design intent. In the AI era, that gap silently widens.10 min read·nomuraya.hashnode.dev
AAman Mulani·Jun 7Agentic TDD Non-NegotiableFor Agents, TDD Isn't Optional7 min read·tech.amanmulani.com1
MMat Banik·May 14Vetting AI as Productivity ToolFour questions developers aren't asking about AI coding tools, and why the answers matter more than the hype7 min read·matbanik.hashnode.dev
