Shiwei Song

Previously @PalantirTech, @Apple, @Stanford, and @CarnegieMellon

[Shiwei Song @shiweiso]

When a model nails something hard, it leaves a good taste in the mouth. I trust it more with everything else. That feeling lasts a while.

[Shiwei Song @shiweiso]

Claude Max plan lasted me just 3 days this week (and that was with lighter usage than before).
Enabled usage credits → set a $50 limit → gone in 10 minutes. Time to give Codex a real shot.

[Shiwei Song @shiweiso]

vibe ported ghostty to windows, probably has lots of edge case issues but passes smoke test https://t.co/LZ8GkkHiVs

[Shiwei Song @shiweiso]

https://t.co/aVxng3aAoI cool project, hard to imagine doing this today without AI

[Shiwei Song @shiweiso]

Jotted down some lessons learned from creating openclaw skills https://t.co/shs2ALmUPR

[Shiwei Song @shiweiso]

Google AI mode prompt injection?

[Jiquan Ngiam @JiquanNgiam]

When we ask an agent to "fix this bug," we're approving hundreds of actions we'll never see. These agents have our production credentials; and increasingly, they run in the background while we work on something else.

Engineers are spinning up multiple Claude Code sessions in parallel, and ClawdBot hit >100k GitHub stars in weeks by letting agents handle email, workflows, even car purchases. The shift to agents without constant supervision is here.

Today we're launching @MintMCP_AI : governance for AI agents.

[Jiquan Ngiam @JiquanNgiam]

Cursor and MintMCP are partnering to help enterprises govern their coding agents through Cursor's new hooks program.

Engineering orgs are rapidly increasing their use of Model Context Protocol in coding workflows. But most orgs have no visibility into what's actually installed, what data flows through these connections, or how to govern them.

Cursor's hooks enable you to observe and control the agent loop by intervening at specific points. With MintMCP, you can enable hooks that fire before and after MCP tools are used.

After enabling these hooks, you get:

🔹 A full inventory of all MCP servers installed across your organization
🔹 Allow/deny policies on servers and tools
🔹 Response scanning for sensitive data before it reaches the AI model
🔹 Complete audit trails for compliance

Start by collecting data to understand actual usage patterns, then layer governance on top.

Check out Cursor's announcement and our blog post on MCP governance using hooks - links in the comments below.

@cursor_ai @MCP_Community @MintMCP_AI

[Shiwei Song @shiweiso]

@PaulYacoubian OpenAI has the worst customer service

[Shiwei Song @shiweiso]

@krishnanrohit Just got this email too no idea what triggered it

[Shiwei Song @shiweiso]

amazing product and demonstration of what's to come in agentic tools! looking forward to the progress ahead

[Manus @ManusAI]

Would you like ask Manus to complete a task for you?

Reply to this tweet with a task you'd like Manus to handle. Our team will run your request with Manus and share the results with you.

Whether it's research, content creation, data analysis, or something creative - let's see what Manus can do!

[Jiquan Ngiam @JiquanNgiam]

AI UX 2024 was a blast 🚀! Meeting @Scobleizer was def a highlight 🥽. We did actual live demos with tasks people came up with and it was so cool to see @Lutra_AI working to solve their problems 🦦

awesome event! @swyx @thesephist @Mappletons @geoffreylitt

[Jiquan Ngiam @JiquanNgiam]

1/ Excited to share more about what we’re crafting at @Lutra_AI ! Lutra generates workflows by transforming your English instructions into code, orchestrating apps to efficiently complete tasks for you.

Our early users have used Lutra for workflows ranging from enriching an Airtable with web data (video below 🎥), web monitoring with sentiment analysis, summarizing Slack conversations to stay on top of things, to automating Gmail responses and Calendar scheduling.

Lutra’s approach is unique: We use AI to generate code that calls out to your apps and glues them together to accomplish it – think code interpreter 🤖 meets apps 🛠️ and integrations.

When it runs, it feels like AI agents that are able to accomplish complex tasks for you. 🚀

[Shiwei Song @shiweiso]

Upgraded from iPhone 14 Pro to 15 Pro Max simply because 14 Pro battery only lasts me half a day. Was battery life always this bad?

[Shiwei Song @shiweiso]

We've been working on Metawork for over a year and launching on Product Hunt today: https://t.co/g81jGHbKh4

Appreciate your support!

[Metawork @GetMetawork]

🎉 Countdown to launch! We're launching Metawork on Product Hunt next week! Gain productivity insights and collaborate remotely via Metawork Virtual Office. Follow our launch and let’s make work better together!
https://t.co/RsYOXiyi24

[Sam Altman @sama]

Move faster. Slowness anywhere justifies slowness everywhere.

2021 instead of 2022. This week instead of next week. Today instead of tomorrow.

Moving fast compounds so much more than people realize.

[Greg Brockman @gdb]

We are going to build something new & it will be incredible.

Initial leadership (more soon): @merettm @sidorszymon @aleks_madry @sama @gdb

The mission continues.

[Sam Altman @sama]

the mission continues

[Satya Nadella @satyanadella]

We remain committed to our partnership with OpenAI and have confidence in our product roadmap, our ability to continue to innovate with everything we announced at Microsoft Ignite, and in continuing to support our customers and partners. We look forward to getting to know Emmett Shear and OAI's new leadership team and working with them. And we’re extremely excited to share the news that Sam Altman and Greg Brockman, together with colleagues, will be joining Microsoft to lead a new advanced AI research team. We look forward to moving quickly to provide them with the resources needed for their success.