Takayuki Kawazoe
Author ProfileClaim This Author Profile
Prove ownership by publishing #HashtagPLUS and this profile link on your author page or an article under your byline. A moderator or admin will review the request before it merges into your real HashtagPLUS username.
🌐 dev.toSource
From Dev.to - ai: "Claude 3, Qwen 6: why we set a different fix_verify retry cap per model"
🌐 dev.toSource
From Dev.to - python: "When 'Control request timeout: initialize' actually means SIGKILL: Claude Code CLI OOM inside Celery"
🌐 dev.toSource
From Dev.to - python: "Cutting MCP token bloat by 12x: what happened when we packed 31 tools into one server"
🌐 dev.toSource
From Dev.to - python: "How one empty message poisoned an entire AI consultation (and the three-layer fix)"
🌐 dev.toSource
From Dev.to - python: "Why your long-running AI agent feels broken (even when it isn't)"
🌐 dev.toSource
From Dev.to - python: "Killing the 5-MCP setup tax with one PyPI package and Device Code Flow"
🌐 dev.toSource
From Dev.to - productivity: How we ended up running one product with 2-3 people, after building our own dev harness
🌐 dev.toSource
From Dev.to - productivity: Codens vs Devin vs Cursor Composer vs Sweep — picking the AI coding agent that matches your bottleneck
🌐 dev.toSource
From Dev.to - python: I shipped a regex to catch AI agents committing only half the work. Then I ripped it back out.
🌐 dev.toSource
From Dev RSS Feed: 3 MCP server failure modes that bit us in production, and how we ship around them