Architecture & Patterns

3+ hour, 2+ min ago   (535+ words) An LLM, or large language model, is a neural network trained on vast text to predict the next token, the engine behind ChatGPT, Gemini, Claude and AI search. "Large" refers to two quantities: the number of parameters, the adjustable numbers…...

57+ min ago   (138+ words) It’s very difficult to talk to an LLM, and you probably don’t even realize you’re doing it wrong. You can instruct an LLM to solve a problem in a certain way, but how good its output is depends on how…...

54+ min ago   (256+ words) People think a single mean value is enough to prove a performance win. In reality, the mean is just one statistic; it can mask outliers, skewed distributions, or systematic noise. When two teams post their means in a chat, the…...

1+ hour, 8+ min ago   (401+ words) Publicado originalmente en iot.gripe, como parte 17 de la serie "Domótica con ESP32 y AWS desde cero". En la parte 3 creamos un certificado a mano y lo pegamos en el código del ESP32. Para un dispositivo está bien. Para diez, o para…...

1+ hour, 8+ min ago   (343+ words) As React applications grow, passing data through multiple components can quickly become difficult to manage. A value that starts in a parent component might need to reach a deeply nested child, forcing every component in between to accept and forward…...

1+ hour ago   (256+ words) Every SaaS company is now building its own agent door. Blockworks shipped a hosted MCP server for its datasets. Notolens built one for brand protection. Genea for access control. An insurer launched an entire directory so every carrier can list…...

1+ day, 17+ hour ago   (537+ words) I hit a wall that felt familiar to anyone building with LLMs. The realization: This isn't an app problem. It's an infrastructure problem. Every team building with LLMs solves the same three things badly: So I built the thing I…...

1+ hour, 6+ min ago   (725+ words) The first time my local agent loop "finished" a task, nothing had changed on disk. The drafter reported success, the critic signed off, the task moved to done - and the working directory was exactly as I'd left it. It took…...

1+ hour, 6+ min ago   (497+ words) Now answer the one question your auditor — or your incident review, or your own 2 a.m. self — is going to ask: Who authorized that? Which human, acting through which agent, with whose permissions? If you're like most teams shipping agents right…...

13+ hour, 33+ min ago   (1608+ words) Imagine you’re using an application and you see a feature you really like. Maybe it’s how Notion handles clipboard operations. Maybe it’s a website with an interesting search implementation. Or maybe you have an old Windows application whose source code…...