by far my favorite detail about this are these icons its a super complex technical feature but there is a lot of sweating the details to make it as delightful as possible kudos to the talented @emtxiao for these

Update: I have joined Notion! After building Mocha, I look at companies and work very differently. I have an extremely high bar which Notion exceeds in multiple aspects. First of all, how many products do you use every day and love using? For me there are roughly 5: my iPhone with its core iOS apps, Notion, and maybe Spotify. I also use AI products (Claude, Grok Bot, Muse...) but those seem to rotate a lot. It feels like a privilege to be able to work on something that's so special to me and so many others. Notion’s mission resonates strongly with my personal ethos, and has a lot of common ground with what I was building at Mocha: “Empowering people to build beautiful tools for their life’s work” through the medium of malleable software. Yep, sounds pretty awesome to me! Our tools are changing faster than ever, mostly in exciting ways but sometimes in scary ways. At Notion they (we!!) believe our tools shape us, and we should build tools which keep humans collaborating and thinking together. I agree very strongly and want to make sure we build a good future. Notion’s team is so stacked! I had a good opinion of them before talking with them, but after speaking with them I was surprised. Every single person I interacted with impressed me, and made me want to build stuff with them. It’s rare to have such talent density. I’m so excited to work with @geoffreylitt @mschoening @brian @nbaschez @colebemis and many others that I can’t mention here. The company is very founder-led and intentional. Sometimes when a company is 13 years old, the founders are checked out. Not here. @ivanhzhao is deeply involved in the product and company, @akothari is constantly trying to improve how Notion can be better explained to the world, and @simonlast has consistently been building at the tip of the AI frontier since before chatGPT came out. A lot of founders sort of “luck into” their domain. If they lived 10 parallel lives, they wouldn’t build the same company 10 times. But some founders would, and I believe Ivan is one of these founders. This means the culture is very intentional: how people communicate, how they work together, how the office is decorated, how they build product, etc… This means that if you are a good fit for the culture, you’re a really good fit. I think it's why so many ex-founders end up here and love it. Finally and importantly: I think Notion is in an incredibly underrated position in the AI race. There is so much competition at the provider level, and although I love the products I feel little to no affiliation or loyalty to model providers. Notion on the other hand is like the comfy data backpack i bring with me when i go from grok bot to muse to instinct, from claude code to codex, etc… It’s the “AI switzerland” which gives people and companies freedom and power to own their own context layer. I think this matters a lot! I’m grateful and honored to be brought into this special team. I’m staying close to my roots and will be building AI products. We're already shipping some big improvements to the harness and custom agents, and there are many very cool improvements coming. I'll be sure to share what we ship along the way, so follow along! Please share feedback and tell me how you use Notion, what we can improve, what you wish it did better, and I’ll do my best to help make it happen!

i personally do not vibe with the muse avatar design feels really weird, like a teletubby that's trying super hard to be neutral big branding miss imo, grok bots are 10x better

The LangChain podcast where @hwchase17 interviews agent builders is full of alpha. Recent one with @EnoReyes was the best one yet. Sharing my unstructured notes: Eno keeps bringing back some core concepts: signals (events that trigger things), workflows (context, skills, and harness config) and missions (the task to achieve). Factory builds on top of these concepts. An example workflow could be code review and an example trigger could be a message in a Slack channel, or a PR being pushed. > The goal of a software factory is to build the workflows which go from signals to deployed code and add as much automation as possible. They discuss memory, a very over-loaded term. There are now many more terms: config, skills, files, wiki, context...? Strong consensus that memory is your moat ( so you really want to keep access to it and own it as an enterprise. Key point of disagreement: Eno thinks wikis are for humans to stay in the understanding loop, while Harrison thinks wikis are for agents to gather context better. Maybe we need two names? Indeed both of these make sense but their shape and requirements will diverge (or will it?). They discuss verifiability. This keeps coming up everywhere in technique discussions for agent building. Once again Eno champions that being able to verifiably check a task with deterministic tests is *the* way to raise the ceiling and achieve long horizon tasks. This includes computer use, code coverage, unit testing, integration testing, evals, ... Factory's missions have a *very* specific set of verification tests. He says they spend a lot of time on these and make them quite exhaustive and specific, and that's a huge determinant in the success of the mission. Speaking of evals, Factory do a lot of them. They don't need to be large! Small "Regression tests" with an N of 10 still hold value (examples he gave were compaction or caching checks). What's interesting to me is that these sound like tests, and the word "eval" has outgrown its initial specific meaning. It used to be a very ML thing with F2 scores and given metrics and properties, and now it captures anything that has to do with testing around LLMs. A lot of human time is spent doing these (example of the guy writing more and more complex lint rules). Of course they discuss how product and engineering roles are evolving. Eno is in the camp that engineers are splitting in two: product eng and deep systems eng. I would argue this was already the case, great engineers typically are one or the other archetype. In enterprise, costs are starting to become an important conversation. This is bullish for two things: open models and model routers. There's this psychological trick where if you do a blind test on users on which model is best you get one distribution, but if you tell them what the models are a lot of preference flows towards the big models from the big labs. Marketing? Eno has a hot take of the [platonic representation hypothesis]( representations in AI models are converging independently of the architecture. In other words, intelligence is discovered and distillation unpreventable, since we end up in the same place and therefore we can borrow the destination (distill) to get there faster. Building a model-agnostic harness. Is this a disadvantage or an advantage? Actually not at all obvious as they showcase quite well. The massive advantage of owning the model is that you can RL your harness, but the disadvantages also exist: you can't mix and match models and leverage the competition, you have biases in your tool call mechanisms (example given is how to apply a diff) which might work slightly better in some areas than others. In the past, models were very prompt sensitive (adding "Don't stop until you're done" changed an OpenAI benchmark by over 30%...), nowadays the models are all getting much stronger which means less arcane word magic and therefore more transferability of prompts between the models. Eno has a fun analogy: preferences aren't optimal. So being left handed isn't better than right handed. You might do stuff much better with your left hand but your performance on the end task is nearly identical. Same thing with how a model would implement a thing (like using a CLI for searching or another, how to apply a diff, etc...). A tactical point Eno makes is that cache-breaking is often worth it, and he sees this as a harness problem, *not* a gateway problem. I suspect this is still an open question and we'll see more experimentation and data play out in the next 6 months (lots of talking about gateways recently).