Linear's Slack bot slammed as "GPT-3.5-level" — pick your team's AI by daily reliability, not leaderboards
sujingshen · x · 2026-10-29
A user complains that Linear's Slack workspace bot is "extremely stupid — seems like they are running GPT-3.5 under the hood." sujingshen draws the lesson: a dumb Slack-embedded bot makes the whole collaboration awkward, and teams picking AI tools should first check whether the assistant they actually open every day is reliable, rather than chasing leaderboards.
More from coding & agent
- PcGöz MCP server gives Claude text-based Windows desktop vision, ~30x fewer tokens than screenshots — emperorbug46 · 2026-10-29
- Code Talks talk: shipping offline in-browser AI with Transformers.js and WebGPU — nicodotdev · 2026-10-29
- AgentWeb launches an action layer that lets AI agents skip browser emulation entirely — rohanpaul_ai · 2026-10-29
- Developer's agent babysits Codex Cloud threads overnight while he sleeps — gabrielchua · 2026-10-29
- Fortune 100 firms still shun coding agents — but engineers have likely smuggled them in already — chrisalbon · 2026-10-29
- Vercel's open-source coding agent harness fx is now a built-in provider in T3 Code — moeinteractive · 2026-10-29