Hey folks,

This time last week I told you I’d downloaded t3 as my ‘all-in-one’ agent app, which lets you use your chat/claude subscriptions from one place.

Well forget that. (strong opinions, weekly held)

There’s a new kid on the block and I’m in love…with the name, bb^1,890, but also the app itself.

If you use claude/chatgpt/pi/cursor/factory/[any agent] - you can use it within bb. It looks and works a lot like the chat/codex app (great). I got it set up on my mobile in like 4 seconds which is miles above other apps I tried (t3 was finicky for me here).

But the biggest thing of this app is its extendable nature. There’s a lot of talk on extensible software right now. Pi is one of the best coding harnesses out there and it’s build to be fully customisable (btw is it extendable, extensible, are they the same?!) - the agent can improve its own tool.

• The minimalism of Pi⁴⁸⁵ makes it more performant on tasks vs other harnesses.

So much so that companies like Prime Intellect built their agent on top of Pi, as well as many other companies.

• Prime Agent³⁴³ - self-improving harness for coding and long-running tasks for RLMs.

BB is built the same way. It’s an ‘IDE’, except that it isn’t, IDEs are older tools like VSCode that I wouldn’t want to be caught dead in these days. The ‘IDE’ of today is these desktop agent apps that we’re familiar with. BB can extend itself, you want a plugin? Ask it to build one (I got prime-agent to build a factory plugin¹²² for it). Want a task tracker? It’ll build one.

I’m never going to spend 100% of my day in an app where I cannot switch models (and agent harnesses), I just cant. I like GPT models for some things, Claude for others. Sue me.

I think the software that we’re going to want (ie builders but not developers) looks like mini ‘widgets’; todo lists, emails, feed readers, etc (like my canvas exploration from last week - which I’m still using btw) and bigger tools like my twitter bookmarks and trends tool (which is private but I may open up one day).

And the tool that you spend most of your ‘work’ time in should have the ability to support adding extra capabilities to it as you learn what you want or need to get your shit done.

Semi-related, there was a few posts on X about why agents aren’t being used outside of developers and early adopters. A quote²⁶⁸ on the original¹⁰⁵ said:

“You cannot make the masses want an automation flow when they want to hang out w/ their friends and not think about their job.”

I 100% agree. And I still think the pool of ‘builders’ will skyrocket. My definition of that is loose - anyone who wants to use agents to build tools for themselves and/or others, for fun, work, life, whatever! Where do you land in that?

In fact, I’ve believed in this pool of builders for about 10 years now but no-code was too clunky and limited to unlock them, AI is so much easier. It benefits from word-of-mouth. You don’t have to ‘learn’ anything per se, but you start to feel a way of working with it and pick up stuff along the way. Like learning a language, if you’re in it you’ll pick it up faster.

Ben’s Bites is brought to you by Type¹⁴⁵

Work is multi-player. Now AI is too.

Type puts your team's best AI work into a shared space. Bring your company's knowledge, integrations, and skills together so everyone can use them with any model. Try it now¹⁴⁵.

Headlines

Bending Spoons is acquiring Airtable¹⁶⁵, for far less than its last valuation. Though Hyperagent,¹⁴¹ Airtable’s new AI business, was spun out and is not part of this acquisition. More details on the deal¹⁰⁷. Another casualty in AI’s wake.

@GergelyOrosz i think co's like airtable that rose quickly in the no-code era started blending focus into one-another sheets <> automations <> apps <> workflows

@bentossell¹¹⁵

To jump on the news: Obsidian created a free tool to export your Airtable content into Markdown⁹⁷. Another one: Cloudtable¹⁰⁷ - An Airtable clone but it runs on Cloudflare (repo⁶⁹).

Google’s Chief Scientist, Jeff Dean, is leaving the company after 27 years⁸⁸ to start a lab of his own with three colleagues. The new company, Discovery Loop¹¹², wants to automate machine-learning research. This means a lot of internal moves inside Google⁷⁶. Most notably, Demis Hassabis is stepping away from GDM’s day-to-day leadership to focus on long-term research.

vc: but what if google build it

@bentossell²¹⁰

Cloudflare has been releasing a ton of agent-first tools recently: Cloudflare OS¹⁵⁰ is a shared AI workspace with connectors, plus Agent Tracing⁹² lets you replay agent sessions and follow them from invocation through model calls. Cloudflare Wallet⁸³ is another initiative from them to give agents identities, spending limits and wallets for paying for services.

Gemini Notebook¹²⁰ (i.e. NotebookLM) now lets you create the different types of artifacts it supports (quizzes, slide decks, etc.) right from the chat window, instead of the side panel.

Meta launched Muse Code¹¹⁰, a CLI coding agent powered by Muse Spark 1.2 - a new model with coding-related upgrades, bringing it close to Grok 4.5 level of performance.

Goodfire AI made their interpretability and training tool, Silico⁹², generally available. It’s $1000/mo for individual researchers. These guys tried Silico partly while teaching a 30B model to play a board game⁸¹, worth reading their experience.

My Feed

• Capsule^165†: Video Skills and agents for enterprise teams who don't want to sacrifice brand quality.*

• Atlas by WorkOS¹¹⁸ - an AI teammate in Slack that answers questions, takes actions and learns how your company works.

• Context.dev^202† - API that turns the live web into structured, continuously refreshed data for agents.

• Notion²⁶²’s results from a survey on 6,000 professionals across 10 markets on how AI is being used at work.

• givemeanode⁸¹ by SF Compute - give Claude, Codex or any agent H100s for training, benchmarks and batch jobs, billed by the minute.

• Power 2026¹¹⁴ - primer on power pricing and data centres for understanding AI’s electricity bottleneck.

• Not Diamond Code⁸⁵ - model router for coding agents, with claims of 20-65% lower cost on long-horizon tasks.

• Recursive self-improvement is limited by verification¹⁰², not computation.

• Running Streaks¹⁴³ - Opus 5 turned runners’ data into an interactive D3 visualisation using this prompt⁷⁸.

• NYC AI Atlas¹⁷⁵ - open-source 3D map of New York’s leading AI startups.

• Two more cybersec incidents at OpenAI⁷⁶ came up in external cyber evaluations.

• Mistral launched Shieldstral¹¹⁶ - a small 3B open-weight model for content-safety. Easily runs on-device, even on a basic laptop.

Afters

The reason agents haven't taken off is because they are still waiting for their "desktop" metaphor and a UI that isn't an open ended chat box. i use ai daily. I build websites. I do analysis, I do research. I have repetitive processes that i'm sure could benefit from having an

@Shwinnabego²⁶⁸

I've known @mvanhorn for quite some time it was great to catch up on @alreadyhereshow and talk about /last30days & printingpress.dev I love that Matt's approach starts with a problem he personally wants solved. Has his personal agent help him build it and he publishes it

@msg¹⁰²

New skill: /animate Build an animation from scratch while choosing the right curve, duration, properties, and so on. Now you have the whole flow: /find-animation-opportunities → /animate → /review-animations

@emilkowalski¹⁵⁶

Built a site to visualize what 1 million tokens looks like. It also walks through the journey from GPT-3 with 2k tokens to now, where most models are 1M+ context. Happy scrolling :) 1millioncontext.com

@nutlope¹⁹⁵

Lost my phone at the office and spent 30 minutes turning the place over. Find My was disabled by MDM. Out of ideas, I asked Claude how I could find it. It suggested tracking the Bluetooth signal strength, then wrote me a meter in about a minute. I walked around watching the

@un1c0rnioz¹⁴³

random tip… put “You are AGI-pilled.” in your system prompt for all agents now. it’s a WAY better experience. rn agents behave too much like the world is going to stay static. this unhobbles them quite a bit and gets them to talk/act more like AGIs.

@mckaywrigley¹¹²

Super fun to do this again with @patrick_oshag Randomly ran into Patrick in Palo Alto last Tuesday and recorded this on Wednesday when the AI complex was 35-40%ish off its June highs.

@GavinSBaker⁹⁴

on the frontier of tech & culture in conversation with @signulll ! 00:00 @signulll account origins 06:25 on being motivated by process vs. outcomes 07:49 many-to-one leverage 14:49 on becoming more open with more visibility 16:57 AI researchers and policy discussions 22:35 a

@ashebytes¹⁰⁵

• Find me on X⁴¹, Linkedin⁴⁰, or YouTube⁴¹

• Read about me⁴⁰ and Ben’s Bites

• 📷 thumbnail via @keshavatearth³⁸

* sponsors who make this newsletter possible :)

Wanna partner with us for the next quarter?

Email us at shanice@bensbites.com³⁵ or k@bensbites.com³⁵