ksapp.eth

@levelsio Not all apps on your device use https, like the iPhone mail app uses IMAP and SMTP (unsecured) unless you force each one to communicate via SSL (IMAPS, etc)

It's called HTTPS

To MCP or not to MCP, that's the question. Lmk in comments

@im_roy_lee Congrats 🎉 You graduated!!

BRO IM ON FUCKING FIRESHIPPPPPPPP

@levelsio @im_roy_lee Now that Leetcode is destroyed… back to in-person tech interviews and white boarding🤦🏾‍♂️

NOoO yOU wOn'T eVer Get a JOB AnYwHEre agAiN!!! yoU WEnT TOO fAr!!! YoU ShoUld ReSpect The iNStiTuTIons AnD NPCs!!!1111oneone @im_roy_lee:

@levelsio Next time buy a Microsoft Surface laptop. You’re comparing 🍎 to 🍊 in terms of price point and quality hardware.

It's 2 hours later and I'm still busy setting up Windows in their wizard Absolutely terribly slow experience with way too many questions Thank god for Apple MacBooks

@jowettbrendan @grok where is the link to the model?

A billion-parameter voice AI is now open source Sesame AI just dropped CSM-1B - the voice engine behind its viral assistant Maya. - Fully open. - Apache 2.0 license. - No paywalls. No permission needed. Here's everything you need to know:

This Claude 3.7 UI workflow is awesome And many vibe coders not aware of it: 1. Have Claude 3.7 to design whole UI screen by screen 2. Ask it to make a plan, break down into components 3. Turn into actual project By having Claude focus on just UI generation, it does a great job You can build beautiful UI without shadcn If you want to learn more about vibe coding beyond basics, join my community here: link.agent.rocks/JNscc5 (also get free access to @10xCoderDev)

@georgepickett @aksmckenzie @DavidSHolz The license for this model is for research only/ non-commercial use

@aksmckenzie @DavidSHolz Has anyone productized it?

can someone make a realtime voice-to-voice language translation ai yet? pls? what are we waiting for?

@aksmckenzie @DavidSHolz The license says that the model is for non-commercial use only- including derivatives of the model. There’s no point in building a product using this model unless it’s for research

@DavidSHolz facebook did! github.com/facebookresear…

starting an online business is dirt cheap: - chatgpt: $0 - next.js: $0 - domain: $10 - resend: $0 (up to 3k emails/month) - stripe: 2.9% per transaction - vercel: $20/month you don’t need investors—just your skills, time, & focus. with consistent effort, you can build something big.

I had the same question but then again it’s a beta release

Why did Google bury its incredible image generation feature in a UX like this? To use it, I have to: 1. Somehow know to use AI Studio instead of regular Gemini 2. Somehow know to pick Gemini 2.0 Flash (image gen) model on the right 3. Somehow know that I can edit images directly using text prompts with this feature I don't get it.

Here’s all the code you need to create a powerful agent that can create and edit any file on your computer using the new text_editor tool in the Anthropic API.

Goodbye Photoshop It's only been 2 days since Google dropped Gemini Flash Experimental, and people are going crazy over it. 13 wild examples so far (Don't miss the 5th one)

This guy literally breaks down how to use AI to write great code

We're launching new tools to help developers build reliable and powerful AI agents. 🤖🔧 Timestamps: 01:54 Web search 02:41 File search 03:22 Computer use 04:07 Responses API 10:17 Agents SDK

@vikvang1 @RobinhoodApp @perplexity_ai @_opencv_ @grok Is this legal? If so, as an alternative could he have used Claude or ChatGPT API without using OpenCV to accomplish the same task of OCR and search? If not, what are other alternative solutions to answering the live trivia contest questions in a similar manner?

i won the @RobinhoodApp $1M trivia contest after building with @perplexity_ai + @_opencv_ turn your volume down near the end

AI is moving so fast that you can start doing new things and become the world expert in them. Just start.

A big realization I've had using a lot of ChatGPT Voice Mode is that we've all trained ourselves to be intellectually constipated by default. It's almost impossible for me to ask Voice Mode a question and then listen to the answer without asking yet another question, recursively. Basically like an annoying child who asks "why" on an infinite loop. Today I paused an audiobook to ask Voice Mode about the Copenhagen Revolution, and 45 minutes later the unbroken chain of questions and answers got us talking about how the Parsi people of India are the descendants of the world's oldest monotheistic religion. It's not that talking to Voice Mode makes me more intellectually curious than when I talk to other humans. It's just that I suddenly have PERMISSION to follow my curiosity wherever it leads. No embarrassment, no fear of being pushy or annoying or repetitive or politically incorrect. Just pure, playful, childlike learning. Turns out we don't "grow out of it" so much as we learn to suppress it. And now we don't have to anymore.

@0xAneri @CoinbaseWallet Have a tool/service that can audit and interact with blockchain transactions to prevent fraud - reducing crypto scams and bad contracts. Let me join your team so we can iron out the details 😀

How can @CoinbaseWallet better support AI agents and those interacting with them. What are the key flows we could make easier?