buggles

@CanesDavid Are these ones you curate or create or default from OE?

Dot phrases sneakily come in to OE "Dot Flows"

@kate_rouch Highly proven the best form of learning is individualized. We’ll see both learning and evaluation change substantially to the benefit of flourishing

The dream of personalized education is starting to be real

@karrisaarinen Is this just a phase of theater for people to pretend to supervise the work

Turns out agents and humans work similar ways

@rauchg @andrewqu Love it. I could see a flavor of the future where people and companies host their own portable plugin ecosystems of skills and workflows they carry around

@_buggles Yeah and more comprehensive than just skills cc @andrewqu

1️⃣ Install the Vercel plugin for your favorite coding agent 2️⃣ There’s no step two. You just gave Claude Code & Cursor production deployment superpowers Running 𝚗𝚙𝚡 𝚙𝚕𝚞𝚐𝚒𝚗𝚜 𝚊𝚍𝚍 gets you every Skill and keeps them updated. So slick

@cuysheffield Honestly a future where access to inference is capital this makes a lot of sense

Request for product - come build this with Visa

@alexkehr Unhinged. I love it

i built an app called Taste where you “lick” the screen to save designs you like 👅 it came from thinking a lot about why tools like Codex feel like they have bad taste maybe the problem isn’t bad taste maybe it’s not knowing your taste (buildwithtaste.com)

@cjpedregal Transcripts with least logical header metadata in markdown accessible via CLI and local storage

There are some tweets out there saying that Granola is trying to lock down access to your data. Tldr; we are actually trying to become more open, not closed. We’re launching a public API next week to complement our MCP. Read on for context. A couple months ago, we noticed that some folks had reversed engineered our local cache so they could access their meeting data. Our cache was not built for this (it can change at any point), so we launched our MCP to serve this need. The MCP gives full access to your notes and transcripts (all time for paid users, time restricted for free users). MCP usage has exploded since launch, so we felt good about it. A week ago, we updated how we store data in our cache and broke the workarounds. This is on us. Stupidly, we thought we had solved these use cases well enough with our MCP. We’ve now learned that while MCPs are great for connecting to tools like Claude or chatGPT, they don’t meet your needs for agents running locally or for data export / pipeline work. So we’re going to fix this for you ASAP. First, we’ll launch a public API next week to make it easier for you to pull your data. Second, we’ll figure out how to make Granola work better for agents running locally. Whether that’s expanding our MCP, launching a CLI, a local API, etc. The industry is moving quickly here, so we’d appreciate your suggestions. We want Granola data to be accessible and useful wherever you need it. Stay tuned.

@jessethanley Can’t wait to try this out

Got early access to this. It's very good and well worth the price. Kinda one of those no brainer purchases if you do any FE development. I think Adam is gonna make a lot of money off this.

It’s not rapid prototyping. It’s not vibe coding. It’s working software

@ow honestly looks great! whatd you use to build?

it’s actually nuts how good AI has gotten at writing code i always lose receipts, so I wanted a tool to track them + manuals, especially for large purchases started playing… 36 hours later I have a solid backend and a decent native iOS app(!!) already. it extracts info from receipts with an LLM, grabs the PDF of the manual + a nice product photo from the internet and takes care of it all. ✨ just crazy how fast you can just make things real end-to-end now.

@nikunj @lennysan this is awesome. I've been riffing on a few ideas of assembling panels/small councils to give feedback on thinking and topics. this is such a good way to get to source perspective for it.

Introducing Lenny's Roundtable 🔥 @lennysan published all his content (videos + newsletters) as a repo. So, I built a website where you can ask it a question and it pulls in relevant people who debate each other - grounded in that content. Live at lennysroundtable dot com 🎉

@amorriscode @ANoffsinger would vastly prefer its one app but with better threading between contexts where chat can branch into code or cowork and back. want to be able to have do a "sidebar" convo to jump out and back in while building total context

@ANoffsinger there’s a distribution advantage to keeping everything together. in the future it’ll be a single tab.

Love the Claude desktop app. I think it's ready to be broken into a couple desktop apps, or simplified

@kevinroose ✌️

If you’re an engineer (or vibecoder) who uses billions of AI tokens a week, DM me. I want to ask you something.

@dillon_mulroy The incantation is the outcome now blake.ist/posts/software…

if you’ve generated a SPEC․md, PRD․md, or PLAN․md and fed it to an agent in the past week, this is mandatory reading so many nuggets of wisdom in here

@ryancarson @linear Have you found you’re not getting close enough with the duct tape version?

The company that actually builds the agent-first code factory is going to be worth hundreds of billions of dollars. No one has cracked it yet. It can't be the model labs because then you're tied to one model. I'm hoping a company like @linear will do this. I'd happily pay thousands of dollars a month for that (+ the token cost). Basically, we need SDLC 2.0 for the agent age. (Also, the right solution can't rely on gh - we need whoever does this to completely replace it as well.)

@jxnlco 👋

How many people are using OpenAI Symphony?

Everyone talking K-shaped and Barbells

@mds I’ve been avoiding it because it essentially means a whole other app for one purpose

Just switched from imessage to telegram for openclaw if you haven’t do it imessage is fine but telegram is so much better

@danshipper Rotate in more agents. Just don’t fry your brain

working with agents is so addicting im finding i have to do reverse pomodoros: work with them for 5 minutes and take a 45 minute break while they run. otherwise im just staring at a thinking trace the whole day