
Building @SerpKite — Google search API for AI agents. Previously built @shipaitoday. Sharing AI, automations, and building in public.
Joined February 2018
- Tweets292
- Following298
- Followers55
- Likes129
Pinned Tweet
I've been building SerpKite: a Google search API for AI agents.
The idea is simple: get live search results into your agent as JSON or Markdown, with MCP and SDKs to plug it in.
It's live now. Try it free—I'd love to hear what you build with it.
serpkite.com
That fun and cool, go and destroy your desktop
We loved @HugoDuprez's Destroy Any Website, so we made one for the Mac.
Open Slack, Jira or your inbox and blow it apart letter by letter.
Destroy Any Desktop is out today.
Find out that T3 Code is surprisingly good
Changed my opinion about @theo
Hey, check my new project, that can be usefull for your ai agents
Meet SerpKite 🪁
The Google SERP API built for AI agents.
→ Live results as JSON or Markdown
→ MCP + TypeScript, Python & Go SDKs
→ Pay as you go. Credits never expire.
→ Failed searches refunded
Start building for free. No credit card.
serpkite.com
Been on GPT-6.1-Sol all day and it's painfully slow. did they throttle inference on purpose to push people onto fast, or am I missing something?
@OpenAI
Nurlan Ibraev retweeted
Your Mac setup shouldn’t stop at one wallpaper.
Walyro lets every display run its own live wallpaper — from calm cinematic landscapes to animated sci-fi worlds.
See your Mac differently.
walyro.com
That was i talking for. I’m also certain they have no agreement with Moonshot AI, and they’re simply hoping there won’t be a lawsuit since the company is based in China. This is capitalism in its purest form: the world’s most advanced and fastest-growing company is using the open-source work of leading engineers and scientists without giving them any credit or paying any royalties.
Yep, Composer 2 started from an open-source base! We will do full pretraining in the future.
Only ~1/4 of the compute spent on the final model came from the base, the rest is from our training. This is why evals are very different.
And yes, we are following the license through our inference partner terms.
🚨 BIG AI DRAMA ALERT 🚨
Cursor just dropped "Composer 2" claiming it's their frontier coding model built with magic RL...
But devs caught the receipts: the internal model ID is literally
kimi-k2p5-rl-0317-s515-fast
→ Composer 2 = Kimi K2.5 + RL post-training
Moonshot AI is PISSED 😤
"Cursor: magic RL ✨ Moonshot: that's our Kimi after RL and stolen tokenizer lmao" (@Avimanyu8008)
xitter.cf/Avimanyu8008/status/20…
"A Kimi researcher confirmed then deleted. Probably secret deal?" (@shantanugoel)
xitter.cf/shantanugoel/status/20…
Chinese media already calling it piracy vibes 🍉
xitter.cf/BlockBeatsAsia/status/…
We live in world where you can fork open source ide, then fork open source llm and worth 50b$. Where did we go wrong?
Supabase is just a well-packaged Postgres stack with ui.
Neon is actually tech changing how Postgres works — branching, storage/compute split, serverless.
Don’t understand anyone who use supabase
There are now 7M developers using @supabase
Signups have been accelerating since the start of the year
Our growth rate right now is as fast as it was during YC, except that we are doing it from a base of millions of developers instead of thousands











