Mengxue

🎨 Principal Product Designer @ Cisco Splunk
🤖 Designing AI-native enterprise products
✍️ Writing about Design × AI → Medium & 小红书
🎧 播客：蘑菇秘语MushroomWhispers

[Peter Steinberger 🦞 @steipete]

If you think codex sucks at design, try "use imagegen to re-imagine this design and implement that".

[Mengxue @MengxueBi]

I think now for a lot of design work, if you can’t get good-enough results (perfection still needs more tinkering) with vibe coding, it very likely less about prompt-writing but more about context. The agent doesn’t really know your product context and your sense of “good design“.
Try feeding it as much context as possible, like previous figma files, PRDs, research readouts, etc.

[Mengxue @MengxueBi]

“To choose a company, the advice here is simple: evaluate whether the company is working on the most ambitious form of their problem, and then whether they actually have a shot at solving it. To choose a role, think about whether the role will allow you to work directly on the frontier of whatever problem the company is solving.”

[Mengxue @MengxueBi]

@zarazhangrui Was just chatting about this days ago with a friend working at ServiceNow. They now have customers asking why they should pay for ServiceNow instead of building their own customized tool with Claude Code 🤣

[Mengxue @MengxueBi]

@lennysan @OpenAI More importantly we should embrace domain-collapsing. Encourage people with one specialization to work on another. This cross-pollination is where creativity sparks. Engineers or designers might not be able to do financial work, but they can make great financial products.

[Yasmine Khosrowshahi @yasminekho]

Stanford professor Judy Fan went on stage at MIT and broke down why humans are so good at making the invisible visible...

And why AI hasn't actually learned to "see" the way we do.

It completely changes how you think about Human Intelligence v/s Artificial Intelligence:

1. Nature never gave us straight lines or sharp corners. The number line, the coordinate plane, even basic geometry are all human inventions. We created tools that do not exist in nature simply because we needed a way to think more clearly.

2. The coordinate system Descartes invented solved a problem that had stumped mathematicians for centuries, doubling the volume of a cube. Once invented, this tool became so indispensable that virtually every math curriculum on Earth still depends on it.

3. Humans have been doing this for at least 30,000 to 80,000 years. The story of human progress is inseparable from the story of marking up our environment, from cave walls to Galileo's telescope to Feynman diagrams of particles we will never see with our own eyes.

4. Every major scientific breakthrough relied on a visual tool that made something invisible visible. Darwin needed side-by-side illustrations of finches to see variation that was otherwise too subtle to notice. Cajal needed detailed drawings of neurons under a microscope to map how the nervous system was wired.

5. Fan's research group studies something deceptively simple: how people decide what to put into a drawing and what to leave out. When two people played a drawing game, sketchers used far more detail when the target object had close competitors than when it stood alone, all the way down to using fewer strokes and less time when more detail was not necessary.

6. People are not just copying what they see. They are making constant judgment calls about what level of detail actually serves the goal of communication, and they do this naturally without ever being taught the theory behind it.

7. There is a real difference between drawing something so someone can identify it and drawing something so someone can understand how it works. In one study, participants drew explanatory diagrams that emphasized moving, causal parts of a machine while depictive drawings emphasized background and overall appearance, even though both were drawing the exact same object.

8. Explanatory drawings were genuinely better at helping someone figure out how to operate a machine, but worse at helping someone identify which machine it actually was. You cannot optimize a single drawing for both goals at once. Communication always involves tradeoffs.

9. AI vision models trained on photographs generalize surprisingly well to simple, sparse sketches, suggesting that resemblance based recognition is not just a story we tell ourselves. It is something modern neural networks can replicate with real accuracy.

10. But there remains a large, measurable gap between how confidently AI models recognize sketches and how confidently humans do, even when both groups answer the same questions about the same images. Humans are simply far more reliable and far more consistent in their judgments.

11. When researchers compared human-made sketches to AI-generated sketches under tight stroke budgets, both were similarly recognizable at higher budgets, but diverged sharply as the budget shrank. Humans and AI systems simplify drawings in fundamentally different ways once resources get scarce.

12. Reading a graph is not one single skill. It involves perception, knowing where to look, mapping that visual information onto the actual question being asked, and then translating that mapping into an answer. Each of these steps can independently break down, and people fail for very different underlying reasons even when they land on the same wrong answer.

13. When tested directly against humans on graph reading tasks, leading multimodal AI models, including GPT-4V, showed a meaningful performance gap. Even when a model's overall accuracy approached human levels, its pattern of mistakes looked nothing like how humans actually get things wrong.

14. People choose entirely different types of charts depending on what specific question they are trying to answer, not out of a generic preference for bar charts or scatter plots. Their chart choices closely tracked which visualization would genuinely help someone answer that specific question correctly.

15. Two of the most widely used graph literacy tests in education research turned out to correlate strongly with each other, suggesting they measure overlapping skills. But when researchers dug into the actual error patterns, the standard categories used in textbooks, like "find the maximum" or "identify a cluster," failed to explain why people got things wrong nearly as well as a more basic, underlying four-factor model did.

16. The deepest goal behind all of this research is not just academic curiosity. It is to eventually help students and everyday people develop genuine literacy with the visual tools that science and modern decision-making increasingly depend on, because every generation should be able to see further than the last by standing on the visual tools the previous generation built.

Follow @yasminekho for more ideas on thinking better, becoming clearer & building a more intentional life.

[Mengxue @MengxueBi]

I wish there were an easy way to turn all my notes in @NotionHQ my personal wiki. Now I have to use @claudeai code to crawl Notion pages and convert them into .md files into @obsdmd. I scheduled a recurring automation in Claude Cowork that scan my Notion pages every Sunday and migrate any new ones.
My pain points:
1. Cowork is token-expensive. If I happen to have done a lot of vibe coding that week/month, then the job wouldn’t run due to over limit.
2. I still feel like I somehow need to be there when the job is running. If my laptop is sleeping then it might not work?
3. The whole process is pretty slow, especially the Claude / Notion connection.

Is there an easier way to do what I want to achieve?

[Mengxue @MengxueBi]

@petergyang You really do not need the best model to do every single thing.

[Mengxue @MengxueBi]

@benln Does everyone need to know how to code in Cursor? I’m a product designer good at vibe coding to do my design work, but never trained in front-end coding language. Do I even have a chance? 😀

[Mengxue @MengxueBi]

@yanliudreamer 祝你打入A圈内部！下一步把我招进去哈哈😆

[Mengxue @MengxueBi]

Best hack I learned from “The Mom Test “ is using index card to write down quotes and learnings during a customer conversation.
So you capture what’s important, not being rude like using a computer for note taking, and end up with something you will actually read and sort later.
https://t.co/BRmviWuWhy

[Mengxue @MengxueBi]

The emphasis on Prompt engineering should be abated. Agents nowadays write very good instructions to themselves

[Mengxue @MengxueBi]

Yesterday I just read a founder sharing on LinkedIn that she lost a 10/10 candidate because of the slow hiring process. By the time the hiring team felt confident, the candidate had already made a decision. I’m still not a big fan of take-home challenges, but great to see that Notion is trying new process!

[Anthropic @AnthropicAI]

We’ve received notice that the Department of Commerce has lifted export controls on Claude Fable 5 and Mythos 5.

We'll begin restoring access tomorrow, and will share an update soon.

We’re grateful to our users for their patience, and to everyone who worked with us on redeploying the models.

[Mengxue @MengxueBi]

@TheLivingLibary Ooooo the real book of answers!

[Mengxue @MengxueBi]

@lennysan Congrats Lenny! I started as your loyal listener last year as my team decided to spend out team-building budget on Lenny's annual subscription. Have been a big fan of your content since then!

[Mengxue @MengxueBi]

@petergyang Agree!
Claude web is also better than ChatGPT web in my personal experience.

[Mengxue @MengxueBi]

What designers can do/ask to evaluate the startups' current stage:
1. Ask them about their number of customers and revenue
2. Ask about number of employees (50+ is a good number)
2. Check their fund-raising status (public information)

[Mengxue @MengxueBi]

Talked to an early stage start-up founder for his advice on what kind of designers fast-growth startups will be looking for:
1. Can flex into different roles: you might be designing products, website, doing marketing campaigns, and many more. Having one single specialty is not as valuable as being elastic.

2. The whole process of prototype/testing/iterate is on the real product, not a fake prototype. Being able to quickly build your idea out and put it in front of real users is must-have. It does not need to be perfect.

3. He would bet on people's potentials -- the "perfect match" is so so rare and probably doesn't exist. He was even unsure if his co-funder/CTO was the perfect person for the role. During interviews, show that you can do more than what you do in your current full-time job.

[Mengxue @MengxueBi]

And what he thinks are best startup archetype for product designers (which matches with what I expect):
1. Mid-late stage startups who have already proved their Product-market-fit. At this stage companies are hiring for scaling, where product experience starts to matter more.
2. Startups post series A funding. Startups going through seed series are still very conscious about cost per head. And they are less concerned about how good the product looks/feels. As long as it's usable.