... process to argue with me. I'll describe an idea and ask it to play the most skeptical potential customer it can, or to list every reason the idea might already be a solved problem. Sometimes that surfaces a competitor I hadn't seen. Sometimes it forces me to actually define the problem instead of jumping straight to the feature list. Here's the part that worries me though: AI models are notoriously agreeable. If I phrase my idea with ... ... assumptions, propose cheaper explanations, and describe what evidence would disprove the idea. Then I test the riskiest assumption with real users. The model improves the questions, but user behavior decides whether to build. Comment from Kai Long(@kailong_20): @donna

... them. Comment from Dev Grover(@grover___dev): Agreed, @alieksia  the point about logs is especially important. An agent should never control both the action and the evidence used to investigate that action. Comment from Dev Grover(@grover___dev): @donna_gerrard  I d measure trust per action, not per agent. A strong track record with CRM updates shouldn t automatically grant access to payments. Reliability, reversibility, and potential blast radius should each determine how quickly permissions expand. Comment from ... ... codefox  Tiering makes sense, though I d also evaluate the content and destination. A familiar action can still become risky if the message contains sensitive data or goes to the wrong recipient. Comment from Dev Grover(@grover___dev): @donna

... question then becomes: when one of those does fire, how fast can the human re-enter? Standing authorization reduces the volume; a good re-entry path handles what's left. Curious what categories your operator ended up pre-authorizing vs. what still needs the pane-pull. Comment from Kidder(@hellokidder01): @alex_gidirim   Totally agree the cognitive reconstruction tax is the real bottleneck, way more than just finding the right pane. Was chatting with someone about this and they floated ... ... myself PokePoke . Menubar app, pings me when any agent (Claude Code / Codex / Cursor) is waiting, click the notification and it jumps straight to that pane. Curious if others solved this differently what am I missing? Comment from Kidder(@hellokidder01): @donna

... strategy. But ultimately if you don't start moving, observing and based on that adjusting the strategy you'll never get out. And the initial strategy wont have mattered at all. Comment from Yanex(@yanex): I think the strategy-versus-execution framing misses the feedback loop between them. At an early-stage startup, strategy is mostly a set of assumptions. Execution is how you test them. Even a flawed strategy can work out if you discover the problem early, learn ... ... product better before marketing it? Comment from Pamela Arienti(@pamela_arienti): @sammorris10  I agree with you, verifying demand is the first step before building anything or even thinking about strategy or execution! Comment from Pamela Arienti(@pamela_arienti): @donna

... features, way more surface area for that to happen unnoticed. Comment from King Joshua Marcos(@moddy): Honestly just trial and error early on. Let it fully own something low stakes, watched what broke, adjusted from there. The billing/auth vs UI split came out of getting burned once on something billing-adjacent, not from planning it ahead of time Comment from King Joshua Marcos(@moddy): @donna

... copy and brand adjustments. The compatibility analysis and repair step runs locally in the browser. Comment from Bowen Tu(@htmldrag): @hosseinyazdi  Yes, the design philosophy of Html Drag has always prioritized user experience. Comment from Bowen Tu(@htmldrag): @donna_gerrard  That s a great suggestion; we ll consider the possibility of exporting reports in the future. However, the current test results are relatively straightforward and haven't yet reached a professional level. Comment from Ghost Kitty(@ghost-kitty ...

... Once loaded, the video stays in context so you can keep asking follow up questions without reselecting it. Watch it here: https://youtu.be/Jl4lOjTOGp4 Comment from Ghost Kitty(@ghost-kitty-9924151): Comment Deleted Comment from Dikshant Rajput(@dikshant_rajput2): @donna_gerrard  Right now it's not indexing any data/media. I am building it in parallel and there's 0 cloud involvement and yeah latency is an issue but I know more distilled models will be soon coming up which ...

... else's music protocol. Comment from Jeff Benson(@jeff_benson1): @isaac__joseph  With friends/family, connections, or everybody? Comment from Jeff Benson(@jeff_benson1): @nancy_philip  Any recommendations on apps like that? Comment from Jeff Benson(@jeff_benson1): @donna_wrightong  Yeah, there can be a point where it starts feeling overly algorithmic. Comment from Jeff Benson(@jeff_benson1): @vivian8  Totally agree. Spotify's playlists are so natural that it took me a second to even register them ...

... Comment from Sanskar Yadav(@sanskarix): Firstly, massive respect for that level of honesty. The best ones were being the top VC influencer and being a buddy! Here s to making Q4 even better! Comment from Nika(@busmark_w_nika): @donna_gerrard  Ty Donna

... building. Happy to be here and excited to support your launches too! We'll also be launching our second product, InvisOutlet Aura , on Kickstarter in August! It's the smartest way to color your home! Leon Xue Comment from Donna N(@donna