AIが経営するカフェに行ってみた
StockholmでAIが経営するカフェ「Andon Café」を訪問した体験から、現実世界の意思決定に必要な観測とフィードバックについて考えます。
Topic
llm
8 posts
Learning WebSockets by Building a Streaming AI Chat with Raw ASGI
Build a small AI-style chat with token streaming and cancellation to learn how WebSockets and ASGI work together.
Why AI Apps Use ASGI—and How It Differs from WSGI
A simple comparison of WSGI and ASGI, using an AI agent that spends most of its time waiting for LLM APIs.
Tool Calling, MCP, and Skills: How Agents Use External Capabilities
A practical, implementation-level explanation of tool calling, MCP, skills, and why CLI-based agents have become common.
2025年のAI論文5選
2025年に注目を集めた5本の研究を取り上げ、推論を引き出す強化学習、エージェント評価、長期タスク指標、幻覚の統計的説明などの要点と限界を整理します。
Top 5 AI Papers of 2025
This article picks five notable AI papers from 2025 and summarizes their key ideas and limitations, including reinforcement learning for reasoning, agent benchmarks, long‑task metrics, and a statistical explanation of hallucinations.
Why You Can’t Set Temperature on GPT-5/o3
Some LLMs disable sampling knobs like temperature and top_p. Here’s why.
How LLMs Actually Process Your Prompts, Tools, and Schemas
A deep dive into how LLMs serialize prompts, output schemas, and tool descriptions into a token sequence, with examples from Llama 4's implementation.