https://scholion.thluiz.com/tags/llms/Recent content in Llms on ScholionHugo -- gohugo.iopt-BR© 2026Mon, 05 Oct 2026 18:13:23 +0100https://scholion.thluiz.com/notes/how-ai-does-and-does-not-change-the-way-i-do-math/Mon, 05 Oct 2026 16:19:16 +0100https://scholion.thluiz.com/notes/how-ai-does-and-does-not-change-the-way-i-do-math/Rachel Webb argues that LLMs will change how she executes math research, but not what makes it interesting. She frames math as both personal enjoyment and a community practice.https://scholion.thluiz.com/notes/how-i-use-llms-as-a-staff-engineer/Fri, 25 Sep 2026 22:47:22 +0100https://scholion.thluiz.com/notes/how-i-use-llms-as-a-staff-engineer/The author uses LLMs as smart autocomplete, a tutor for new domains, a fallback for bugs, and a proofreading tool, while avoiding them for full drafts and familiar production work.https://scholion.thluiz.com/notes/the-fate-of-small-open-source/Fri, 25 Sep 2026 18:25:20 +0100https://scholion.thluiz.com/notes/the-fate-of-small-open-source/The post argues that LLMs make small utility libraries less necessary, which weakens both their reason to exist and their role in teaching developers. Open source still matters, but the author thinks its value is shifting toward larger, more inventive, or niche work.https://scholion.thluiz.com/notes/llms-bring-new-nature-of-abstraction/Fri, 25 Sep 2026 17:31:55 +0100https://scholion.thluiz.com/notes/llms-bring-new-nature-of-abstraction/The article argues that LLMs change software development as much as the move from assembler to high-level languages, but also add non-determinism, so programming becomes a different kind of interaction with the machine.https://scholion.thluiz.com/notes/llms-are-mirrors-of-operator-skill/Fri, 25 Sep 2026 08:52:31 +0100https://scholion.thluiz.com/notes/llms-are-mirrors-of-operator-skill/The author argues that AI is a matter of skill: LLMs reflect the operator’s level, which makes it harder to identify qualified operators and shows that the interview process is broken.https://scholion.thluiz.com/notes/llms-will-lie-forever/Fri, 25 Sep 2026 00:37:49 +0100https://scholion.thluiz.com/notes/llms-will-lie-forever/The text presents an article that argues that hallucinations in LLMs are inevitable, based on mathematical arguments, and that this should change how we think about reliability, fact-checking, and the use of AI.https://scholion.thluiz.com/notes/llms-reward-expertise/Wed, 23 Sep 2026 18:06:45 +0100https://scholion.thluiz.com/notes/llms-reward-expertise/The post argues that the main skill in prompting is domain knowledge. Expert users can steer LLMs harder, spot odd answers, and get more useful results than people who only know general prompting tips.https://scholion.thluiz.com/notes/llm-coding-workflow-2026/Wed, 23 Sep 2026 15:59:28 +0100https://scholion.thluiz.com/notes/llm-coding-workflow-2026/Addy Osmani lays out the workflow he converged on after a year of AI-assisted coding: spec before code, small chunks, heavy context packing, and treating every AI commit as his own responsibility.https://scholion.thluiz.com/notes/a-primer-on-large-language-models-and-their-limitations/Wed, 23 Sep 2026 15:38:15 +0100https://scholion.thluiz.com/notes/a-primer-on-large-language-models-and-their-limitations/Johnson and Hyland-Wood’s LLM primer argues hallucination is the wrong word: LLMs produce Frankfurtian bullshit, indifferent to truth, alongside catastrophic forgetting, model collapse, and jailbreak risk.https://scholion.thluiz.com/notes/news-feature-what-are-the-limits-of-deep-learning/Wed, 23 Sep 2026 15:38:15 +0100https://scholion.thluiz.com/notes/news-feature-what-are-the-limits-of-deep-learning/PNAS profile of 2019 surveys where deep learning breaks: adversarial spoofing, data inefficiency, opacity, and the missing common sense that symbolic AI once tried to hand-code.https://scholion.thluiz.com/notes/after-two-years-of-vibecoding-back-to-writing-by-hand/Wed, 23 Sep 2026 14:35:35 +0100https://scholion.thluiz.com/notes/after-two-years-of-vibecoding-back-to-writing-by-hand/Mo Bitar traces two years of AI coding through excitement, spec docs, and what he calls slop, and explains why he went back to writing most of his code by hand.https://scholion.thluiz.com/notes/code-is-cheap-show-me-the-talk/Wed, 23 Sep 2026 13:52:07 +0100https://scholion.thluiz.com/notes/code-is-cheap-show-me-the-talk/Kailash Nadh argues LLMs made code abundant and cheap, so accountability, provenance, and the ability to think and articulate now matter more than the code itself.https://scholion.thluiz.com/notes/i-dont-like-llms/Wed, 23 Sep 2026 12:25:47 +0100https://scholion.thluiz.com/notes/i-dont-like-llms/Fowler feels both fascination and fear about AI, but talking to LLMs directly leaves him with one dominant reaction: dislike, traced mainly to the Silicon Valley culture that shaped these models.https://scholion.thluiz.com/notes/how-i-use-ai-oct-2025-ben-stolovitz/Tue, 11 Nov 2025 15:39:13 +0000https://scholion.thluiz.com/notes/how-i-use-ai-oct-2025-ben-stolovitz/Ben Stolovitz describes where AI helps him most: coding autocomplete, research for “pub facts,” summaries, transcription, and editing. He is skeptical of AI search, writing, and image generation.https://scholion.thluiz.com/notes/llms-are-not-like-you-and-me-and-never-will-be/Tue, 02 Sep 2025 15:23:38 +0100https://scholion.thluiz.com/notes/llms-are-not-like-you-and-me-and-never-will-be/The post argues that LLMs are not human-like minds but autocomplete systems that lack proper world models, so they keep making obvious errors in time, history, economics, and factual reasoning.https://scholion.thluiz.com/notes/why-your-ai-never-works-on-the-first-try/Tue, 12 Aug 2025 17:36:30 +0100https://scholion.thluiz.com/notes/why-your-ai-never-works-on-the-first-try/The article argues that AI output often needs many iterations because LLMs only move step by step through possibility space. That makes AI a cognitive exoskeleton that needs human navigation.https://scholion.thluiz.com/notes/the-case-against-conversational-interfaces/Thu, 03 Apr 2025 09:28:54 -0300https://scholion.thluiz.com/notes/the-case-against-conversational-interfaces/The essay argues that conversational interfaces are too slow and lossy to replace existing computing tools, but can work as a complementary layer that adds voice-driven actions across apps and workflows.