Bookmarks

VoidZero, the team behind Vite, Vitest, Rolldown, Oxc, and Vite+, is joining Cloudflare. Vite stays open source, vendor-agnostic, and built for everyone.

The June 2026 AI Gateway production index: DeepSeek's token share jumped to 17% as low-cost models entered production, while Anthropic held 65% of all spend.

The June 2026 AI Gateway production index: DeepSeek's token share jumped to 17% as low-cost models entered production, while Anthropic held 65% of all spend.

Feature flags offer a practical way to manage this risk by separating the act of deploying code from the act of releasing a feature to users. Instead of a single, high-risk launch event that affects all users, teams can ship code to production with new features hidden by default.

A markdown viewer for the terminal that renders images and Big Headers™ - benjajaja/mdfried

はじめに こんにちは、MA部SREブロックの片桐です。MA部ではメルマガやLINE、アプリプッシュ通知を配信するためのマーケティングオートメーションシステムを開発・運用しています。 MA部ではDBとして主にCloud SQL for MySQLを利用しており、調査や不具合対応のために開発メンバーがDBにログインして各種…

Behavioral differences and prompting patterns for Claude Fable 5 and Claude Mythos 5, covering effort, instruction following, long runs, memory, and scaffolding changes.

【Web担】「ドメインパワーが高いサイトほどSEOに有利」は本当なのか。SEO業界の第一人者・住太陽氏に聞いた。

X(旧 Twitter)、GitHub のようなハンドルネームのあるタイプのサービスでは、アカウント登録時に誰にも使用されていないハンドルネームを決める必要があると思います。この際、以下のように入力と同時にハンドルネームの使用可否が分かると

はじめに こんにちは。基幹システム本部 基幹開発部 商品管理ブロックの田中秀明です。 Claude CodeやCodexの利用が広がるほど、各人の使い方、プロンプト、レビュー観点、AIへ任せる範囲がばらつき始めました。AIを高度に使いこなせる人は開発の進め方そのものを変えられる一方で、これから使い始める人にとっては「ど…

Next-Gen AI-Aware Git. Auditable sandbox, model/prompt-aware commits, 95% less token waste, 3.5x richer PR brief, 1.8x faster multi-agent loops. - h5i-dev/h5i

AI Engineering Summit Tokyo 2026の登壇資料です。 価格.comは、C#とClassic ASPが混在するWindowsシステムを、創業から約30年、抜本的な刷新をせずに運用してきました。コードは約960万行、テーブルは13,210に及びます。この積み重なった技術的負…

Today we’re launching Claude Fable 5: a Mythos-class model that we’ve made safe for general use.

Today’s coding benchmarks have established that models can write correct code, but the question we should really be asking is: can models actually write good code?

Amazon Bedrock AgentCore Runtime gives each agent session its own isolated microVM with a persistent workspace, secure tool access through Gateway, and built-in observability—so you can run Claude Code, Codex, Kiro, and Cursor in parallel without sharing secrets, ports, or filesystems. Close the lid, go to dinner, and pick up where you left off tomorrow.

Stream Postgres to Apache Iceberg on S3 via logical replication, queryable over the Postgres wire protocol. - viggy28/streambed

Keep your castle secure and moving.

secrets-protecting agent router and model gateway focused on security and convenience as primary goals - bglusman/calciforge

Some of the tech giants for the past few years have tried hard to kill the most important tool of the web, links. The web is changing and folks are losing hope. I just wanted to thank those who made the internet interesting for me.

松江Ruby会議12 登壇資料

A cron job for every man, woman, and child.

A Python framework for self-hosted LLM tool-calling and multi-step agentic workflows - antoinezambelli/forge

AI is using AI to use AI to do my job! What do I do now?

Zero-config, fast `io_uring`-based HTTPS server. Contribute to losfair/zeroserve development by creating an account on GitHub.

A local-first desktop IDE for Amazon ECS. Contribute to utibeabasi6/mercek development by creating an account on GitHub.

A Rails-flavored guide to wrapping Claude Code in the checks, conventions, and feedback loops that make agent output more trustworthy.

Technical debt lives in your code. Cognitive debt lives in your head. Intent debt lives in the artifacts you may have never wrote: the goals, constraints, an...

macOS ちょっと前から、そういえば Apple インテリジェンスなるものが使えるなぁ、ただなんかただ邪魔なUIが出るだけだが（自分のユースケースでは全く使わない）、程度に思っていたのだけど、macOS のネイティブAPIとして呼び出せるようなので、それを使って翻訳 cli ア...