Ben Karciauskas

Trying to be a better husband and dad every single day.

[SpaceXAI @SpaceXAI]

Grok Voice Think Fast 2.0 is now #1 on the Artificial Analysis Speech-to-Speech Index.

This index measures whether voice agents can reason over the speech it hears, resolve real customer issues, and correctly complete tasks using agent tools.

[lauren @poteto]

something i really love about working on @cursor_ai and @bot is that we just love to ship. i think it shows up across everything we do, like not building products and models that get in your way. we dogfood our own work obsessively and aim to be the number 1 power users of our own tools.

and that's why @grok is built for people who love to ship too. give us a try if you're on the fence!

[Kun Chen @kunchenguid]

ok i rabbit holed and ended up creating a full-on SOFTWARE FACTORY within Grok @Bot.. and it works

i call it Grok Ship - a ship you'll captain, and it ships! yesterday alone, hundreds of issues and PRs across my repos got done by it

no bs - sharing my full setup here 👇

[lauren @poteto]

github is down, time to switch to Origin!

[Lee Robinson @leerob]

Grok @Bot has made a few simple yet powerful technical decisions that I believe make it easy and enjoyable to use.

1. The best UI is none at all. The product interface is dramatically simpler than alternatives without sacrificing functionality. How is this possible? It's one of the first products designed for current frontier model capabilities and has a UI restrained enough to remain easy to use as models improve exponentially. Everyone knows how to text.

2. A thin harness for the client, a thick harness for the server. You might have noticed the app feels very fluid to use, even for a beta product. This is primarily because of everything we didn't have to build. The app harness is essentially a single tool to send messages between the client and server. The complexity moves to the server, where you can still use the coding agent harness with specialized tools as needed. This helps make the UI fast and responsive on desktop and mobile.

3. An always-on computer. Most coding agents and assistants today start fresh with every question you ask. Some of these sessions are on your local machine and others happen in the cloud. We believe strongly that cloud is the future, which is why it's the only option. Further, rather than spinning up virtual machines for every conversation, your bots connect to their own computer. This means you can still run agents on the bot's persistent filesystem. It's closer to what programmers have been doing by using Tailscale from their phones to connect to a remote computer and run an agent TUI. You get those capabilities without the hassle.

4. Your bots can use the browser. Coding agents have shown that most work on a computer can be expressed and run as code. You can ask for a task in natural language and the agent will decide to write a script to complete it. This is amazing, but there's still many tasks which can't be completed without logging into a website and clicking around the browser. Models and harnesses are now good enough to reliably handle this. The combination of writing code and using browsers means you can automate almost any task on a computer. Further, you can ask Grok Bot to record you doing the task, and then turn it into something repeatable.

[Kun Chen @kunchenguid]

i'm 80% sure Grok Bot was originally built by the Cursor product team, and got rebranded after the acquisition

a few tells:

1. the iOS app is published by Anysphere, not X Corp. who published the Grok app

2. the mac app download URL is hosted under cursor dot com

3. it seems to run on cursor's vm infrastructure

4. it's cursor team members who are actively responding on X about the topic

in either case, this is a solid release - good work!

[SpaceXAI @SpaceXAI]

Introducing Grok 4.6.

It delivers frontier intelligence and is a significant improvement over Grok 4.5 at the same price.

[Cursor @cursor_ai]

Cursor Router keeps improving from millions of in-product user interactions each week.

We intelligently classify and route requests, lowering latency and reducing cost based on the task.

[Cursor @cursor_ai]

Cloud agents are now 20-30% more token efficient, and 80% more efficient on runs with computer use.

We've improved how they handle MCPs, skills, and computer use, so you can delegate more ambitious tasks and get back demos while staying within budget.

[lauren @poteto]

if you haven't already setup @cursor_ai cloud agents, i highly recommend it! these days my standard workflow is to have one local coordinator agent that spawns cloud subagents (https://t.co/S0OcjqYIBi) to do all the work. the coordinator is in charge of managing everything and ensuring my subagents follow my pstack playbooks. if you haven't tried it yet, use /in-cloud to spawn a cloud subagent!

it also works extremely well with /swarm (https://t.co/0z0KqFz3Xi), a new skill i made that spawns a whole bunch of cheap and fast subagents, each with their own vm and computer, to do anything you wish.

my favorite use case is to spawn a swarm of 10 or more Grok 4.5 subagents to run our app in the cloud and fuzz it to find all sorts of problems with PRs before i ship them. when you combine this with my autopilot playbook, you can basically ship PRs in your sleep.

[NVIDIA @nvidia]

Read why AI agents need more than one model. 👇 https://t.co/IFaesvTdl8

[Lee Robinson @leerob]

We just doubled the included usage of Cursor models on all plans.

Enjoy more access to Grok 4.5 and Composer 2.5!

[Jordan Topoleski @jordantop]

Cursor Start is live in India!

Our new plan gives developers generous access to Cursor's models for coding, local and cloud agents, and the ability extend and customize Cursor with plugins, MCPs, rules, and hooks. Give it a try!

[lauren @poteto]

https://t.co/6Ta0f4J3u2

[Xiaoyin Qu @quxiaoyin]

Kimi just surpasses Fable across multiple benchmarks, and it’s open weight! China has caught up.

[Cursor @cursor_ai]

We're launching code storage and git hosting.

Origin gives teams and agents a place to host, review, and collaborate on code.

Available this fall. Join the waitlist.

https://t.co/uamaIarJXY

[Caspar @caspar_br]

Great stat in here: Claude Code went from 17% to 92% on our eval set once it had access to LangSmith traces and Skills. A coding agent without trace data is just guessing at fixes

[Andrej Karpathy @karpathy]

Software horror: litellm PyPI supply chain attack.

Simple `pip install litellm` was enough to exfiltrate SSH keys, AWS/GCP/Azure creds, Kubernetes configs, git credentials, env vars (all your API keys), shell history, crypto wallets, SSL private keys, CI/CD secrets, database passwords.

LiteLLM itself has 97 million downloads per month which is already terrible, but much worse, the contagion spreads to any project that depends on litellm. For example, if you did `pip install dspy` (which depended on litellm>=1.64.0), you'd also be pwnd. Same for any other large project that depended on litellm.

Afaict the poisoned version was up for only less than ~1 hour. The attack had a bug which led to its discovery - Callum McMahon was using an MCP plugin inside Cursor that pulled in litellm as a transitive dependency. When litellm 1.82.8 installed, their machine ran out of RAM and crashed. So if the attacker didn't vibe code this attack it could have been undetected for many days or weeks.

Supply chain attacks like this are basically the scariest thing imaginable in modern software. Every time you install any depedency you could be pulling in a poisoned package anywhere deep inside its entire depedency tree. This is especially risky with large projects that might have lots and lots of dependencies. The credentials that do get stolen in each attack can then be used to take over more accounts and compromise more packages.

Classical software engineering would have you believe that dependencies are good (we're building pyramids from bricks), but imo this has to be re-evaluated, and it's why I've been so growingly averse to them, preferring to use LLMs to "yoink" functionality when it's simple enough and possible.

[Greg Brockman @gdb]

a small window into the opportunity of AGI

[Ben Karciauskas @chowgi]

Cursor's computer use capability is mind-blowing! It wasn't getting the result it wanted by a curl command on the demo I am building and it wondered if it was some sort of websocket issue. It opened the browser and navigated the app and ran the test it needed. Mind blown.