Hey NYC 👋🚆 so excited to share my team's submission to the @MTA open data challenge: subwaystories.nyc! A map of over 1B subway rides

I smell fear

We've reset usage limits for all Grok Bot users. Enjoy!

Over the past week and a half we've made major improvements to make Grok @bot 3x faster. Glad you can feel them!

Grok Bot can now suggest ways to help without you needing to ask.

This has been huge for me. Perhaps I will share some nice charts soon

...

Almost one year ago

Invite a @bot to your team!

When a nerd tells me they don't use agents "because I can write better code than any agent", I usually go "huh, sure buddy, you do you". When they tell they don't use agents for monitoring, I just laugh in their face. Seriously. You can't monitor a production deploy better than an agent can. I dunno about AGI but this specific job is something that an LLM can already do better than every single human in the world. For me, as a systems engineer, having agents monitor my infra changes has been (no joke) more impactful in my day-to-day work than having agents write my code. I started doing this many months ago by asking the agent to monitor explicitly after I triggered a deploy; soon after I moved to using a set of my own Skills. Now our Review Agents team has fully automated this workflow with delightful UX and obviously I'm not going back. It's hard to overstate how powerful this is. All it takes, for me, in production: - The Datadog MCP - The PlanetScale MCP (for all our databases) - Our custom MCP for deploys and internal tooling That's it. Some carefully designed gradual deploy pipelines, and the agent just takes care of everything from there. You can't really deploy SEVs to production anymore. Doing infra like this is pure joy. And the safety is just a tiny part of it. The Rollout agent also does the actually hard part of my job, which is _performance_. When the changes you're deploying are a performance optimization, the agent prepares a monitoring plan that ensures the optimization is correct AND effective. I rarely prompt benchmarks for my optimizations anymore. Just fucking go to prod and the agent is gonna give me the best possible benchmark. If you work on systems or infra for a living, you should be setting up your production environments so you can use Rollouts. Not doing so is actually reckless.

I may be jinxing it but Grok Bot is back. I can tell the Cursor team responds to feedback well. Grok Bot's snappier and they fixed the weekly usage drain

Btw, think this chart from today's blog has so many interesting takeaways: - Large majority of coding agent cost is from cache hits. - Reasoning accounts for a full 10% of cost now. When models overthink it costs a lot - Reading files costs more than other tools combined

they made a movie about HOAs?

I optimized Cursor's agent harness. Check it out for nitty gritty and interesting details on where token spend comes from. This one was actually a tricky rollout because tokens are so important to our top line. But we care about keeping low prices thanks to efficiency.

We heard feedback that @bot was a bad texter - your bots will now reply quicker. Oh, and another 6% more usage on average! As always, expect more wins coming soon.

remember when tech companies would publish their roadmap? sounds so twee now

Forgot to mention that you can talk too.

Update: We've rolled out more optimizations to grok @bot to give you another 10% more effective usage. This round boiled down to reducing overuse of subagents, less noisy bot-to-bot chatter, and smarter routing for low effort batch work. And a pro tip: beware group chats! They are super token guzzlers and frankly I'm not sure the ROI is there for most people. The bigger the group chat, the more it costs. If you're price-sensitive I would avoid these for most work. We might need to limit these until we can bring it under control.

Using AI to hillclimb calling real people in a loop is unethical and should be banned.

Game changer for Grok Bot

BREAKING: NVIDIA CEO Jensen Huang says Grok Bot is incredible. 😎🔥