100% private · no tracking · works offline100% client-side/no data leaves your browser/no accounts/works offline
§ The diary · page 2
81 articles on web development, tooling, React, JavaScript, and practical frontend work.
Page 2 / 9

OpenAI API
A production migration guide for moving recorded and live speech-to-text workloads to GPT Transcribe without losing subtitles, timestamps, speaker labels, or reliable language handling.
Jul 29, 2026· 11 min read

OpenAI API
A production runbook for setting organization and project spend caps, handling insufficient_quota errors, and keeping critical features available when an OpenAI budget is exhausted.
Jul 26, 2026· 11 min read

JavaScript
A current implementation guide to exposing safe, structured website actions to browser agents with WebMCP, including forms, JavaScript tools, lifecycle control, security, and testing.
Jul 23, 2026· 13 min read

AI Agents
A practical migration plan for the stateless MCP release candidate, covering request metadata, explicit state handles, caching, retries, deprecated features, and staged rollout.
Jul 20, 2026· 12 min read
LLM
GitHub Models shuts down for good on July 30, 2026 — this guide walks through finding every place your code depends on it and migrating to Azure AI Foundry, OpenRouter, a direct provider API, or a local model, with working code.
Jul 17, 2026· 11 min read

JavaScript
A production guide to Programmatic Tool Calling in the OpenAI Responses API, including tool eligibility, structured contracts, nested-call continuation, routing boundaries, and operational safeguards.
Jul 17, 2026· 13 min read

JavaScript
A production-minded LiteRT.js guide for running compatible AI models locally in the browser, from conversion and WebGPU execution to fallbacks, memory cleanup, caching, and testing.
Jul 14, 2026· 13 min read
Developer Tools
A step-by-step walkthrough of installing ZCode, configuring Z.ai's open-weight GLM-5.2 model, running Goal Mode and MCP, and calling the API directly — plus how it stacks up against Claude Code, Cursor, and GitHub Copilot.
Jul 11, 2026· 13 min read
LLM
Context engineering, not prompt engineering, is what keeps AI agents coherent past a few turns. This guide covers the four core techniques (write, select, compress, isolate) with runnable Python examples for each.
Jul 10, 2026· 14 min read

AI Agents
A practical breakdown of GitHub Copilot's new AI Credits billing — how tokens convert to dollars, what every plan includes, and how to keep one agent session from eating your month's allowance.
Jul 9, 2026· 13 min read