Pedro Lucas Marques·4d agoOLS: o que os mínimos quadrados podem nos ensinar sobre atenção e memória em LLMs?Da regressão linear à memória episódica: uma aplicação prática no desenvolvimento de agentes inteligentes.
InGarofalo UX Blog: Signature UXbyGarofalo UX·Oct 30The Advent of LLMs in Frontend UX Design: Pros, Cons, and CompetencyHow large language models are reshaping the way we design, develop, and deliver digital experiences …and why human judgment still matters.
InSkill StuffbyTechByRahmat·Oct 28RAG vs Fine-Tuning: 7 Decisions That Shape AI ApplicationsChoosing between RAG and fine-tuning is not just a model decision. It determines how your application learns, retrieves information…A response icon1A response icon1
Abdulrahman Maslmany·Oct 18Model Context Protocol (MCP) for Local LLMs: Connect Ollama & vLLM to Workstation ToolsAn engineering blueprint for air-gapped FastMCP servers, dynamic client schema discovery, and deterministic local tool calling.
Ahmed Adawy | AI Systems & Python·Oct 4Demystifying LLM Serving Infrastructure: How PagedAttention and Continuous Batching Scale InferenceMoving a Large Language Model (LLM) from a local prototype in a Jupyter notebook to a high-throughput, multi-tenant production environment…
Dmitry Mamyrin·Aug 17Software Changed. Our Mental Models Haven’t. Principles for Working With AI.Why outcomes, experimentation, and better mental models matter more than prompts and workflows. Lessons from using, advising, and building.
InTowards AIbyPrashrey Pradeep·Jul 14No, that is not how you prompt!From Zero-Shot to ReAct: A dev-friendly, zero-fluff guide to the 10 prompting frameworks that actually matter.
Parvez Mohammed @ Techlatest.net·Jun 15When to Fine-Tune an LLM (And When Prompting Is Enough)Adapt a pre-trained language model to your task, domain, or behavior without retraining from scratch. This guide maps the modern…
Tayler Erbe·Jun 1The Metrics That Lied to Us: What We Learned Evaluating LLMs for ProductionA performance engineering deep-dive into three open-weight models, structured extraction, and why the “obvious” quality metric was actively…
Javier Alcívar·May 29The Great AI ObfuscationHow Model Tiers, Effort Levels, and Naming Schemes Hide Real Performance
