junu

Founder. Designer. Fullstack Builder. Investor. Lifelong Learner. Maker. @Usercallco Previously @PresenceChat @UserLookco @IDEO

[junu @junetic]

(\ /)
(^.^) chop = claude hop

[junu @junetic]

Switching Claude Code sessions across projects sucks

chop

https://t.co/G1HbQZiL7r

[junu @junetic]

shows all sessions across projects/agents → pick one → jumps straight back in

[junu @junetic]

Taste is knowing which details matter most

[junu @junetic]

micromanaging is toxic for humans

it’s essential for AI

AI doesn’t think
it executes

the difference between average and great is the details you bothered to define

[junu @junetic]

@clarkcharlie03 @thiingsco very nice. instead of static collection I think you should let users generate any icon with a prompt and charge based on usage

[junu @junetic]

@jdreeves Curious. Do you often run customer interviews to get insights on brand positioning, messaging etc to inform the strategy or direction?

[junu @junetic]

@Jibran_05 https://t.co/mTGOsGqATR

[junu @junetic]

@Johnsjawn Show don’t tell

[junu @junetic]

@danshipper Analysis is great start. Next step is let agents talk to users. Even better—let micro user interviews be auto triggered from Posthog events @Usercallco

[junu @junetic]

@sharbel Why would anyone use this when Claude can directly edit your website code and shownthe results in production

[junu @junetic]

@geoffreywoo We jut did it for our user research saas (let AI agents talk to users for you): https://t.co/X2yvTfjqKa

[junu @junetic]

@bnj Yes. In addition to personal taste, judgement and experience, you need real user insights to ground them. AI is lacking on this front but now AI agents can talk to users for you as well https://t.co/X2yvTfjqKa

[junu @junetic]

@larsencc One missing loop is user understanding. Product events → quick user feedback → insight

[junu @junetic]

@DsMatie AI can talk to users for you now so you can take more calls with the right people - https://t.co/XoWBlFkMEj

[junu @junetic]

@AvinashDalvi_ @akshaymarch7 There is an AI agent mcp for talking to real users - https://t.co/XoWBlFkMEj

[junu @junetic]

@DJbennyBuff @SlackHQ @meetgranola @sentry One missing signal is capturing why right when behavior happens. Product events triggering quick user conversations could feed that layer

[junu @junetic]

@signulll And knowing what to build is easy as having ideas. Knowing what to build that users actually need requires talking to them. And now AI is also helping to do that - https://t.co/XoWBlFkMEj

[junu @junetic]

@aakashgupta Knowing what to build is easy as having ideas. Knowing what to build that users need requires talking to them. And now AI is also helping to do that - https://t.co/XoWBlFkMEj

[junu @junetic]

@signulll Feels right. When building gets easier, judgment becomes the bottleneck.

The shared model of “why” also works best when it’s grounded in real user conversations. The signals from what users do and why they say shape both what to build and how to frame it