这印度哥们本想拉中国下水，结果被网友给教育的妥妥滴，评论区一片中国好。
These racists have a lot to say about Indians, but why are they so quiet when it comes to China? Maybe because they know China won’t tolerate the same level of disrespect. They assume Indians are too polite and well-mannered to fight back, so they keep pushing their limits.
Never mistake an Indian’s good manners for weakness. Being civil doesn’t mean being an easy target.
ShowMeOnMap now has its own community in Reddit.
reddit.com/r/ShowMeOnMap/s/D…
What mountains can you see from your next destination? 🏔️
showmeonmap.com/mountain-fin…
I built a little mountain radar map with @claudeai opus 5.5. @ClaudeDevs
Click anywhere on the map to discover the surrounding peaks based on elevation and visibility.
Apartment hunting? Buying a house? Booking your next Airbnb?
Maybe your next place has a mountain view you didn’t know about. 👀
Grok @bot turned my phone line into an AI front desk
The number is just a pipe on @telnyx. @grok Voice answers as my receptionist and speaks whatever language the caller uses. Grok @bot is my butler behind the scenes: it handles callbacks, outbound calls, encrypted logs and daily backups, and texts me friendly summaries.
The fun part is the anti-scam layers:
-Carrier screening and caller attestation
-Behavior scoring, to tell a human from an AI caller
-Covert checks that plant a wrong detail
-Claims must stay consistent
-Money or emergency means hang up and call a known number
The AI never reveals anything. It only checks what the caller says.
Here's my latest architecture #GrokBot #VoiceAI
A big mountain finder feature completed in ShowMeOnMap. If you curious about the location where you can see the mountains around, check this out!
Got my Grok Bot and Claude Code talking to each other on my Mac 🤯
Grok writes the brief, Claude drafts spec + plan, they peer-review each other, I approve from either phone app, Claude builds.
Localhost only, no open ports, every command approved. Diagram below 👇
让 Grok Bot 和 Claude Code 在我 Mac 上直接对话了🤯 Grok 写需求，Claude 出方案，两个 AI 互相评审，我在任一手机 App 点头就开干。只走本机，不开端口，每条命令都要我批准。架构图丢给你的 Grok Bot 就能照着搭👇
Hi Grok @bot, launch drone Delta from dock 4, pick up my order from @PandaExpress, grab a bubble tea in @HEYTEA, and don't forget to check my uncle's cat, make sure they are ok, over.
youtube.com/watch?v=rIznH_5I…
北京工作日的所有城轨列车，从首班到末班 🚇 一共 32 条线，包括地铁、机场线、磁浮/有轨电车和市郊铁路。
灵感来自 @milos_gis 做的瑞士火车动画，用 Grok Bot 做的。
线路来自 OpenStreetMap；列车位置是按首末班和发车间隔推算的示意时刻表（部分为估算），不是真实运行记录。
给Grok Bot搞了它自己的电话(604-216-xxxx)以后，今天给它手下的接线员立了规矩，涵盖日常生活的绝大部分场景，包括识别诈骗，套路等等，接线员现在知道：能说什么，不能说什么，怎么说，明天让Grok Bot安排接线员给牙医诊所打个电话看看会不会直接被对方前台拉黑挂掉哈哈哈哈哈哈哈
FAI retweeted
Agree. Claude was able to do this better than I could since Opus 4.8 or so. 5.5 models are now very good at it.
All you do is give the model
1. a goal (eg. “Make p95 ci time <10 mins”)
2. a way to measure its result (eg. “Read timings in bigquery after each change”)
3. a sense of how hard to try (eg. “Don’t stop till it’s done, it might take a few weeks of experimenting”)
This pretty much works for most things. I often also ask the model to show me a chart with progress to goal every day, or to dm me a live artifact with the result