IDIvan Dimovinivandimov.dev·Feb 13 · 6 min readThe Death of the Flaky Test: Why I Stopped Writing Scripts and Started Architecting AgentsIt’s 2026. If you are still manually updating CSS selectors because a div moved three pixels to the left, you are doing it wrong. For the last decade, we’ve been stuck in a loop of "write, break, fix,00
IDIvan Dimovinivandimov.dev·Feb 8 · 5 min readThe RAG Triad in 2026: Testing with LLM & DeepEvalIt is 2026. GPT-5, DeepSeek V3.2, Gemini 3 pro… are here, and reasoning capabilities are nothing short of extraordinary. But let’s be honest: if your RAG (Retrieval-Augmented Generation) pipeline feed00
IDIvan Dimovinivandimov.dev·Feb 6 · 4 min readThe $47k Loop: Why Your AI Agent Needs a Circuit BreakerThe "Notebook Phase" is the most dangerous place in AI engineering. We’ve all been there. You hack together a prompt, chain a few API calls in a Jupyter notebook, and hit Shift+Enter. The output is magic. You show your PM, they’re thrilled, and you s...00
IDIvan Dimovinivandimov.dev·Feb 5 · 4 min readWhy Your RAG App Is Slow (and how to prove it)You hit "Enter." The loading spinner starts spinning. You wait. You take a sip of coffee. You wait some more. Finally, five seconds later, the LLM spits out an answer. It’s accurate, sure. But in the world of software, five seconds is an eternity. W...00
IDIvan Dimovinivandimov.dev·Feb 4 · 4 min readThe Evaluation Bottleneck: Building a "Golden Dataset" Without Losing Your MindIf I see one more "vibe check" evaluation in a pull request, I’m going to scream. You know the drill. You tweak the prompt, you run a few queries in the playground, it "feels" better, and you merge. Two days later, a user asks a question about a spec...00
