Maxi Lieven

@bridge_surf great job, how did you built the cua system? Built it ur self or using some kind of mcp?

This is one of the most interesting and meaningful use cases we’ve seen recently! Bridge helps users build Shortcuts workflows without requiring any scripting knowledge. Of course, if you don’t want to watch the AI complete the task on your screen, you can also switch to background mode🤖

@ionleu taptlk.com

drop ur startup link

@ogbuildsapps hey lets connect

who's gonna be my lucky 250th follower 🥹 if you're a builder into tech, ai, saas, marketing - say hi 👋 and let's connect 🤝🌱

@strategy_parody @sanketnadhani thanks, appreciate you trying it. and yeah, dictation should feel like it gets more reliable over time, not less. if you hit any issues or have feedback, just reply here, happy to help fast

@maxilvn @sanketnadhani Will try. Disappointed that wisprflow degraded. I mean ML should get better now worse by month.

Back in the market again for a dictation tool. Done with Wispr Flow, done with Monologue, couldn't get Muesli set up. What else?

@thomaskffmn @WisprFlow i use my iPad as an extended display too, and TapTalk still works fine for me there it’s built as a native Mac dictation app, so it keeps typing into the focused Mac app even with Sidecar/external display workflows. happy to onboard 1:1 or you can test it: taptlk.com

Hey @WisprFlow, did you know that using your iPad as an extended display for Macbooks completely disconnects the capabilities of Wispr dictation?

@ShinghiD @WisprFlow model changes are exactly why i wanted TapTalk to run on-device on Mac. same local model every session, no surprise cloud changes, no audio uploads. if Wispr quality keeps shifting on you, happy to onboard 1:1 or you can test it: taptlk.com

Did @WisprFlow change their AI model again? Just went back to garbage mode.

@babablahblah_ @sanketnadhani that Claude-folder setup sounds useful, but the extra step is the friction i’m trying to remove with TapTalk. it runs on-device on Mac and puts your voice directly into email, WhatsApp, Slack, etc. without audio uploads. happy to onboard 1:1 or test it: taptlk.com

@sanketnadhani I have a Claude folder which converts my voice input into email/whatsapp/slack msg.. probably the best solution right now.. it’s a small extra step, but only set up that works for me.. I agree that wisprflow has been a let down :/

@strategy_parody @sanketnadhani if Wispr feels unusable and Muesli was hard to set up, try TapTalk. it’s a native Mac dictation app, runs on-device with no audio uploads, and is built to work directly in whatever app you’re typing in. happy to onboard 1:1 or you can test it: taptlk.com

@sanketnadhani Yea wisprflow is totally useless for me now

@unimaginative taptlk.com

Startup founders, post your startup's website in the replies and I'll review the design of it! 😀

@ryancarson How about your personal agent - still using openclaw with chiefclaw setup?

My current setup: 1. backend eng + automations in cloud = devin $2500 / month 2. design = claude design $200 + $500 credits / month 3. frontend eng local = codex $200 / month

we love hearing real workflow feedback like this. this is how TapTalk gets better: users share what feels missing, and we try to turn it into shipped improvements.

@SLudhiyani @WisprFlow @TapTalkOfficial this is great to hear, thank you for actually testing it. keep sending any issues or improvements you want to see. i’ll keep trying to turn real workflow feedback into shipped features.

@maxilvn @WisprFlow So I've tried this app and it's working pretty well and also I've enabled this feature which you shared it's working pretty fine as well. Okay, so yeah great stuff and this tweet which I am s saying I should say is also by @TapTalkOfficial ⚡️

I'm trying out @WisprFlow for the first time. It's so damn good. Feels 3x faster already.

This guy knows how to cook 🫡 Shared my feedback, and less than hour he added that feature in his product If anyone looking for app like #whisprflow , give a try to @TapTalkOfficial And he is just 17! Keep going ⚡️

@SLudhiyani @TapTalkOfficial really appreciate this. this is how TapTalk gets better: share issues or improvements you want to see, and i’ll try to implement them. love hearing real workflow feedback

@nizzyabi You can also let Codex or Claude mock ui changes in html and tell it what to change, literally so unnecessary to use Figma step before

why can’t i just use an agent to design my app/site in code? i don’t understand the added layer to designing then transferring that into code.

@swerikcodes How did you need 30h, did you code it by hand??

Every once in a while you just have to lock in and get shit done. Rebuilt the entire Meuze landing page in 30 hours, barely slept.

@SLudhiyani @WisprFlow also improved this feature by adding auto-submit apps, so you can choose where auto-submit is active. tell me how it feels in your workflow, happy to hear feedback

@maxilvn @WisprFlow Damn, very impressed with this. Keep going ⚡️

@ryancarson @HelloUntangle Since when are you using codex and not devin anymore?

Locked in. Onboarding first big law firm next week at @HelloUntangle Codex + subagents cranking on a huge PR

@SLudhiyani @WisprFlow Shipped it just now. Not a huge deal hance I have already experimented with the target feature Working atm at a small improvement of this autoenter, guess you will like it ;)

@maxilvn @WisprFlow Did you shipped this feature just now or it exists already in your app.

@SLudhiyani @WisprFlow Update the App and feature will be in TapTalk now

@SLudhiyani @WisprFlow ahh you mean after dictation it should instantly hit enter/send, and that sometimes fails? would you want that as an option in TapTalk too?