Are LLMs still surprisingly bad at some simple tasks?

I Built a Visual System Design Reference — Switch Any Architecture Between AWS, Azure, GCP & Open Source

I Built a Visual System Design Reference — Switch Any Architecture Between AWS, Azure, GCP... Tagged with systemdesign, webdev, opensource, showdev.

MCP Explained

An interactive guide to the Model Context Protocol: every JSON-RPC message on the wire, step by step. Transports, OAuth, tools, resources, sampling, elicitation, tasks, tool search and codemode.

The Web Is More Than Enough, by Paul Ford

Web browsers enable incredibly rich app development—and that’s radically easier than it used to be.

Learn Vector Databases for Beginners – Embeddings, Similarity Search & ChromaDB

A vector database stores data as lists of numbers called embeddings, which lets it search by meaning instead of matching exact keywords. It's the piece behind a lot of recommendation systems, RAG apps

RAG vs. Vectorless RAG, clearly explained!

...explained visually (with tradeoffs in production).

DeepSeek Super-Agent Architecture Explained

How DeepSeek Harness's Super-agent works: a Router, isolated Workers, event triggers and a Watchdog let one AI agent run many tasks. Diagrams inside.

Learn How to Design Software Systems Like a Senior Engineer in Our New Course

Bối cảnh: Thiết kế hệ thống là kỹ năng quan trọng đối với các kỹ sư phần mềm hiện nay, đặc biệt khi ứng tuyển vào vị trí senior hoặc staff-level. Nguyên nhân kỹ thuật: Khóa học mới này tập trung vào các nguyên tắc thiết kế architecture và giải quyết bài toán scale cho hệ thống thực tế. Hệ quả: Học viên sẽ nắm vững cách thiết kế hệ thống như một senior engineer, tăng khả năng ứng tuyển vào các vị trí cao cấp. Điều đáng học: Khóa học bao gồm các case study thực tế với công nghệ như microservices, load balancing và database sharding.

Khóa học này giúp bạn nâng cao kỹ năng thiết kế hệ thống phần mềm như một kỹ sư senior.

The End of AI Frontiers

Large language models remain far less efficient than biological brains at learning, representing knowledge, and planning, the post argues. A child hears roughly one token per second, reaching a billion tokens only after 31 years, while GPT-4-class models reportedly use four orders of magnitude more data. The Phi model series, trained on 10 trillion tokens, still trails GPT-4 along most axes. A Portia spider plans prey-catching behavior with a brain a million times smaller than a human's, and squirrels recall cached food after months, suggesting that existing models may be far less representationally and memory efficient.