techwithleena.hashnode.devWhy Consensus Matters More Than Confidence in AI SystemsWe are building our digital infrastructure on a fault line. The current generation of Large Language Models (LLMs) suffers from a specific, dangerous pathology: they are programmed to be confident, not correct. When you ask an AI a question, it does ...Jan 19·6 min read
techwithleena.hashnode.devThe Failure Boundary Where LLM Reasoning Quietly CollapsesLarge language models feel impressive right up until they do not. The responses still look fluent. The structure still appears logical. But somewhere beneath the surface, reasoning quality drops. Assumptions blur. Constraints leak. The model keeps ta...Jan 16·5 min read
techwithleena.hashnode.devA Production Rule for Handling Model UncertaintyYou are shipping gambling algorithms, not software. I look at the codebases of "AI-native" startups, and I see the same terrifying pattern. A developer makes an API call to an LLM. They get a response. They JSON.parse() it. And they push it to the fr...Jan 15·5 min read
techwithleena.hashnode.devLLMs Fail Less Randomly Than You Think. The Pattern Is the ProblemWe treat Large Language Models (LLMs) like chaos engines. When an LLM hallucinates a library that doesn't exist, or confidently explains a security vulnerability that isn't there, we shrug. "It's just the temperature," we say. "It's the stochastic na...Jan 9·4 min read
techwithleena.hashnode.devDesigning AI-Assisted Systems Without Creating Fragile DependenciesWe are currently witnessing a massive architectural regression in software engineering. For the last twenty years, we spent our careers trying to make systems deterministic. We built unit tests, type safety, and immutable infrastructure to ensure tha...Jan 8·6 min read
