@lawrluk @NotionHelp The way it is supposed to work is it remembers wherever you were scrolled last, so if you reload it should stay in the same place.
I think it is fixed, let me know if you're still seeing an issue!
English
Simon Last
2.2K posts
can someone who works at @NotionHQ explain why every time I refresh a page, the timeline view jumps to this date where nothing is happening, instead of a sensible default like
- today
- when the first visible item would be
I asked your AI and it referred me to a setting that does not exist (:
English
Simon Last retweetledi
AI is making software easier than ever to build. The bottleneck has moved.
Now the hard part isn't "can we build this", it's "should we" and "who needs to know."
That context is scattered across Slack threads, meeting notes, tasks, GitHub.
I've wanted to fix this for a long time by building a software lab. One shared system where people and agents work from the same context. People make the judgment calls. Agents do the triaging, the routing, the summarizing.
Ship OS is that system. This is Notion's first packaged product for the full product development loop... docs, databases, workflows, and agents in one setup. We've run our own launches on it for months. Including this one.
And it's live today: notion.com/ship-os
English
Simon Last retweetledi
The best place to create and share HTML, whether it’s built by Notion Agent or Claude + Codex.
More launches coming very soon! Our goal is to make Notion agent the default cloud agent for knowledge work.
Cloud agent + shared workspace + team context = a new kind of computer.
Notion@NotionHQ
New block in Notion: HTML. Build interactive HTML right on your Notion page. Ask AI to turn your content into interactive explainers, prototypes, or diagrams. Share with your team to use and tinker together.
English
@arag_agrawal Idea is it’s there if they want it, I think if you don’t it will be frustrating for more advanced users. Imagine AWS if it didn’t let you terminate an instance.
English
@simonlast are we sure on #3.b?
if the agent is the primary user, why give humans low level access to the primitives?
English
A few weeks ago, I joined @NotionHQ.
I’ve spent most of my work life designing systems for knowledge and code, so I couldn’t be more excited to help build the future of how it all comes together.
English
Coding agents make this especially obvious. Internally, we have an agent called Boxy with access to our codebase, services, secrets, tests, CI, and deploy paths.
External agents let Boxy run in our own infrastructure, while the collaboration happens on the Notion task.
English
Useful agent work should attach to the task, doc, or project where the work is already being discussed.
Teammates should be able to follow along, comment, review output, see changes, and understand permissions.
English
External agents are coming to Notion.
Claude and Cursor are first. More agents will follow, and we’ll also provide a way for companies to connect agents they’ve built themselves.
Notion@NotionHQ
Introducing External Agents in Notion: Claude + Cursor. Your team already collaborates in Notion. Now your favorite agents do too. Assign them tasks from a board shared with your whole team. @-mention them like teammates. Watch them run.
English
I'm really excited to be working with Engram to bring their models into Custom Agents in Notion. There's something magical that happens when the agent stops needing to search your workspace, and just knows it.
Engram@EngramLab
English
Pretty wild how under-studied women’s health seems to be!
>>>
• From 1977 to 1993, FDA guidance recommended excluding women who could become pregnant from Phase I and early Phase II drug trials.
• 8 of 10 prescription drugs withdrawn from the US market between 1997 and 2000 posed greater health risks for women than men.
• Heart disease is the leading cause of death in women. Yet women are about 50% more likely to be initially misdiagnosed during a heart attack, and those misdiagnosed face about a 70% higher risk of death within 30 days.
• In cardiovascular trials, women are still underrepresented relative to disease prevalence in 62% of studies, with no improvement over recent decades.
• Women receive less pain medication than men at the same reported pain score. In one controlled study, nurses rated an identical case as less painful when the patient was described as female.
• Endometriosis affects roughly 1 in 10 women and takes an average of about 11 years to diagnose in the US.
English
@osama_latrache This should be fixed now! Let me know if you’re still seeing an issue
English
Thanks for digging in, mate.
quick clarification — it's not about formula length, it's about chain depth.
1-hop works fine: prop("X").first().prop("Y") → auto-resolves
2+ hops break: prop("A").first().prop("B").first().prop("C") → comes back as "‣" ❌
So when i ask the agent to apply a multi-hop formula directly (color theme lookups, deep traversals across related dbs), the property refs return as chips and the formula breaks.
workaround i use: i ask the agent to write the formula as text, i paste it manually, then Notion's ui picker injects the chip.
that works. but agent end-to-end fails on multi-hops.
English
@mike64_t If this is happening, you need to set it up so it is more likely to achieve the task. Try creating a plan doc, then iterating on it until you feel confident it makes sense and can be achieved.
English
/goal has to be one of the most useless features to ever exist. Burns tokens and time like a bottomless pit, and whenever you interrupt it and ask what it’s doing it’s perusing some strategy so doomed you have the urge to violently face palm. Usually 10 tailored leading yes or no questions just relaying its own information back at it will prune the branches by like 90%. My condolences to OpenAI serving a feature to ddos themselves.
English
@osama_latrache > breaks long formulas on paste — property references become "‣"
Mind sharing specific steps to reproduce this?
> can't touch workspace settings
What would you like to change about workspace settings?
English
@thdxr I don’t think that’s a good analogy in this case.. the performance of the agent is the quality of the product
English
imagine a benchmark of two editors
one opens a file 10x faster! wow it must be better. but oh wait they just disabled syntax highlighting
real products have to do things that make it worse on benchmarks but better in practice
Ben (no treats)@andersonbcdefg
official confirmation that the claude code harness has become slop btw
English