Register and share your invite link to earn from video plays and referrals.

Geoffrey Litt
@geoffreylitt
malleable software @NotionHQ / prev @inkandswitch, @MIT_CSAIL / 🇯🇵🇺🇸
1.9K Following    25.9K Followers
TI-83 BASIC while bored in class, iykyk Fun story: I made a prank calculator that gives subtly wrong answers, to install on my friends’ calculators… and then accidentally used it myself for a chemistry test
Show more
Be honest: what was your first code editor?
in my experience, agent skills can totally remain relevant over time / across model generations, *as long as they are more spec than step-by-step instructions* eg: specifying a good format for a doc output, a good flow for interviewing a human
Show more
my personal recommendation when people ask me where to put their AI context: docs, skills, memories, etc.
I’ve long been inspired by Lex and Nathan’s other projects from afar. Now I get to be inspired up close!
Update: I’ve joined @NotionHQ! I am ridiculously pumped to build the perfect space for people and agents to think together. What an incredible time to join. The business is ripping, the team is insane (in a good way), and yet so much work remains ahead of us. It’s an honor and a thrill to be here and to help write Notion’s next chapter. Some of you may be wondering, what does this mean for Lex and Roughdraft? Lex will be winding down by the end of the year, while Roughdraft will continue on as an independent open source project. I’ll give a more detailed update soon (especially for Lex users about the transition — I'll make sure you are taken care of), but the main message I want to share today is gratitude. Thank you to @trueventures and all our investors, to the Lex team, to our customers and users, and to friends and family who supported us along the way. Also, a huge thank you to @ivanhzhao, @akothari, @mschoening, and everyone at Notion for bringing me aboard and making me feel at home. Most importantly, thank you to @SoniaBaschez. When we sat down for dinner in August 2022 and you told me you’d support me spending nights and weekends on that crazy AI writing idea I had (with an infant baby girl at home, no less), we had no clue what the next four years would look like. We had no idea we’d get hundreds of thousands of users within weeks, or that Sequoia would put Lex on a market map. We had no idea ChatGPT was coming in a few months, or how AI would reshape the world. We had no idea Lex would become not just be a side quest at Every, but an independent company. All we knew was that we had a good feeling about it, and you were willing to bet on it. Anyway, as you can tell, I am insanely excited for what comes next. Can’t wait to share some of the things we’re cooking up with you soon. And whether you love or hate Notion, I’d love to help make it better for you!
Show more
Did a podcast interview organized by some awesome Stanford researchers! Among other things, discussed one of my favorite topics: once code is free to write, what are the infrastructure changes that need to happen to take full advantage of that democratization?
Show more
“When I say understanding is the new bottleneck, what I mean is that I need to keep a clear understanding of what’s going on in my head to advance the project and be a full creative participant in that project.” "Coding agents make coding more lonely. Can we open up that process to make it inherently collaborative?" For EP7, “Software With Taste,” we welcome @geoffreylitt, design engineer at @NotionHQ, whose work explores malleable software and designing environments for thinking. AI speeds up our work, but is our understanding keeping up? Geoffrey explains the risks of optimizing for speed at the expense of human understanding, why he uses AI to quiz himself during code reviews, and how to develop taste and build tools that help us stay creative, think deeply, and work together. 0:00 - Introduction 0:47 - Design Is More Than Better Animation 3:32 - Malleable Software and Coding Agents 9:06 - Human-Software Interaction is a Full-Stack Problem 12:43 - Getting Users On Board and Deeper 15:53 - The Trap of Just Building More 20:27 - Primitive Design Is a Taste Problem 22:20 - Understanding Is the New Bottleneck 27:22 - How To Develop Taste? 34:49 - Productivity vs. Speed 38:10 - Will Better AI Hurt Human-Human Collaboration? 44:34 - Advice for Designing Better Products
Show more
Notion runs one of the largest CRDT deployments in production, processing millions of operations every minute! We wrote about the data structures we adopted from CRDT research to make our text editor more collaborative, and how we built on those concepts to fit our block-based data model.
Show more
the best part of this launch is that this api is 100% compliant with the agent skills spec these skills are not just markdown! you can put code, reference files, images, spreadsheets, and anything else that might tailor your agents better for their task. ...which means that notion now has a new folder primitive that you can use to manage your accompanying files. this will hopefully be a useful new feature for you that you can use for skills and so much more!
Show more
# the document editor is the next IDE I've talked to a lot of non-engineers about AI skills recently, and it's *fascinating* hearing their perspective. A document with a prompt is now a unit of executable software! Just flip a bit. No need to go through an App Store. Just share the document and you're sharing the program. But then as people get deeper into using skills with their team, for real, they start hitting all the problems of, well... sharing executable software! Wait, how do we know the thing still works? How do we collaborate and make suggestions without messing up the running version? How do we discover useful existing modules? Who controls what ships? If you squint, these are classic software engineering concerns!! But emerging in a totally new medium... What excites me about the current moment is that we get a chance to reinvent tools for solving these problems, but without the baggage of professional software engineering workflows. That's why it's so interesting working at @NotionHQ right now. Working on collaborative document editors is a way to define the programming experience for a huge group of people. Today many of our users have already started using Notion to edit their skills and access them in all their agents. Now our goal is to follow our users deeper and deeper down the rabbit hole, giving ever more powerful tools for sharing and improving skills. Because a document is a program now :D p.s. this also relates to a great 2010 academic paper which coins the term "end-user software engineering". In a nutshell, end-user programmers using tools like spreadsheets end up hitting all the classic needs of Software Engineering, and we can invent new tools to help them out!
Show more
@geoffreylitt How does this connect to evals? If everyone can edit the skills then everyone needs to make sure that they’re not just improving one part but making another part worse
Write agent skills in Notion, then install them into your coding agent with the skills CLI. No Git repository required. $ 𝚗𝚙𝚡 𝚜𝚔𝚒𝚕𝚕𝚜 𝚊𝚍𝚍 <𝚗𝚘𝚝𝚒𝚘𝚗-𝚞𝚛𝚕>
Show more
heads up, skills aren't just for engineers anymore! everyone is using skills, and the whole team needs a place to collaborate on them. not a code hosting platform. some place a bit more familiar.....
Show more
if you care about the craft of writing, this talk is worth its weight in gold
@tarstarr @zebriez The best I’ve seen in this genre, of quite an old “keynote address”
@tarstarr @zebriez The best I’ve seen in this genre, of quite an old “keynote address”
More valuable than ever to convert tacit knowledge to written instructions. Aka, teaching
visual function inspector... rendering inputs and outputs with domain specific GUIs
IMO the essential difficulty of malleable software now boils down to “fork maintenance”. Let’s unpack that a bit. If you want to make your own custom tools or fork existing OSS, the cost of that is headed to zero. But… often we don’t want to start from scratch! We want to remix tools others have created. We want someone else to help us maintain stuff. We want to combine our ideas with others’. Now the moment you have multiple parties both updating the software… you hit gnarly problems! In a nutshell: happens if I modded something and it no longer meshes well with the latest updates from you? This is “fork maintenance”. (Traditionally this would be a centralized model where a “dev” is maintaining the software and you, the “user” might adjust settings or patch things. But even in a more decentralized setting we want to jam with our peers on our tools.) A classic approach here is to expose a stable plugin SDK. Works great because the dev can keep the surface maintained and plugins hopefully compose reasonably…. But only works as long as the plugin surface has what you need. Sometimes you end up hitting walls. If it’s OSS you can also literally fork. Maintaining a fork and merging edits from others is a huge pain. But maybe with enough smart cheap LLMs this is more viable now. You also need to consider data, not just code. Schema migrations are already a pain in the ass with normal software; when every user is doing customizations things get extra gnarly. I don’t think that there’s a perfect solution to this yet. Ideally I’d want something much deeper and broader than a typical plugin API - eg a language or framework designed for layered additive modification (think Harel’s behavioral programming as one reference.) I find it quite poetic that collaboration and working with others is where malleability gets really good but also where it gets hard. Same as many other things in life! For more on this, we go into quite some depth on this topic here:
Show more
I don't like that new tools are elevating a disengaged way of creating, like a boss who's too busy to care ("Just text your bot while out for a jog"). That we can do this now is amazing! But great work happens with our hands in the clay. The same tools can let us get deeper in.
Show more
Our Matic makes me feel the way Apple products used to make me feel. I feel the care at every step of the design. Highly recommend for anyone who needs daily cleaning for a kid or pet
Introducing Cues: Voice & Gesture control for Matic We raised $115M and spent 9 years to make the world’s first intuitive home robot Say you spilled coffee: Point to the spill and say, "Hey Matic, clean this" and it will hear, see, locate in 3D, and go clean on its own. Matic comes with a lot of features: 1. Say "Hey Matic" - it locates your voice, turns, and looks at you 2. Say "Hey Matic, follow me" and start walking. Matic will follow behind 3. Say "Hey Matic, go clean the living room". Since it knows your house map, it navigates and just does it It's so easy, a 5 year old and an 80 year old can use it and it understands 75 different languages. Matic has 8x the airflow, specialised cleaning algorithms for rugs, corners, toekicks, mopping, etc and cleans better than any other robot vacuum. Also keeps improving with software updates. 13,000 families use and love Matic. WIRED magazine gave it a 10/10 (the only hardware to receive this rating in a decade) Buy yours at and if you don't love it after 6 months, we'll give you a full refund. To celebrate our launch, we're cleaning 300 homes with Matic in San Francisco and New York City. Comment "Matic" below, we'll send you the link to sign up and come to your doorstep to clean your home with Matic.
Show more
🪧 what do you love or hate about notion
Took the sign to the streets of SF to talk to our customers
In some ways, malleable software is the *opposite* of ephemeral software Instead of making a new thing every time, use the same thing for everything. Familiar, versatile, universal tools.