5) Screenshot of Ceevee on macos:

I built 36 projects in last 14 months, with AI code. Today, I'm sharing my Coding Docs with you. I use these 7-docs in Cursor/Windsurf to build a knowledge base for AI models. Bookmark this post so you can copy the templates. ↓ 1. Project Requirements Document (PRD)

holy shit GPT-4o image gen is unreal. it can generate UIs too 🤯 designers are not safe. 11 examples:

the corporate speed run guide (top 5 tips): my credentials -- high performer at my company every cycle so far and my co workers always end up loving me both as a person and worker 1. be a an empathetic human despite performing well: I know that it's easy for us to think "I'm just going to shut the fk up and do my work with as little human interaction as possible" and to be honest I've had this mindset in the past. I've come to realize though after observing others that were not thriving that no one wants to work with someone / vouch for someone who has no empathy EVEN IF they are the highest performer in the org. Being shut off from everyone else and just doing your tasks works but it's not optimal for "speed running" your corporate journey i.e., get promoted/move up fast. It sucks but there are politics involved in moving up. Also, it's more fun when you have human connections at work so it's a win win. 2. block off anywhere from 2-4 hours of interrupted work FIRST THING when you log on For me this is the time I get basically all my work done and anything after this is either ad hoc work, helping others, or tackling tech debt on the team. If you aren't doing deep work you for sure cannot be the highest performer you can be thus cannot speed run corporate 3. say no to meetings and delegate tasks I outright decline meetings when it genuinely does not benefit either me or my team. Our time is precious, there's a decline button for a reason. You can often judge whether a meeting is truly necessary or not and for the first half of my career I was too much of a yes man and just accepted every meeting -- wasted so many hours. Also, tasks that someone else can do way faster than you should be delegated to those people. You don't look good if you help someone for 2 hours when someone else on the team could've only spent 30 mins helping out. 4. keep an ongoing document of ALL artifacts and result that you produce This saves you time in the long run when performance reviews come around because you have all the things you need to provide to your manager on what you've worked on but also it feels good seeing what you've accomplished throughout the quarters of the year. Make sure to add links and data to make it hard for anyone to doubt your production of good work. Numbers don't lie. 5. share your knowledge and treat your team as a part of your own success Every time I became an expert at a certain task I created a SOP and shared it with my entire team so that they don't ask me questions and because a productive team is good for you climbing the ladder as well. If your team is dragging along it doesn't matter how good you are, your team becomes an eye sore in the org and it's hard to get cases for promos as opposed to if your entire team was pulling hella weight in the org. That's it for now and remember that your day job is NOT your identity. Touch grass, hang out, and make memories outside of work. A life fulfilled is one where you produce good work but also have good human connections ;) If you want to start a career in ML and are already a technical person drop your email, lots of content in the works right now and y'all will get first access: list.raymondyoo.com

Dippy AI: 60+ accounts & 60M+ views in 3mo Crushon AI: 50+ accounts & 80M+ views & $200K MRR app + 11 different AI gf apps doing the exact same mass faceless TikTok growth hack (huge alpha):

Cursor Tip #8: Add your own documentation in @cursor_ai Save this for later 🧵

(1/6) Introducing a powerful solution for managing and steering multiple AI agents in Cursor 0.43+. Start with cloning .agentic-cursorrules alongside your main repository to enforce domain boundaries and prevent cross-agent conflicts:

philosophers of X, what is the “roadmap” of learning philosophy for someone who is new? i see hella technical roadmaps but no one ever makes non technical roadmaps

How I’d launch a startup if I had 0$ for marketing: 👇

We've added tool use support to the Workbench in the Anthropic Console so you can now test prompts that use tools.

Even if you suck at front end, v0 & Cursor if prompted right give great results. All you need is detailed explanation of your project Use o1/claude to create a markdown file that include: - tech stack - file structure - project description - core functionalities Place all this in your root project folder and give it a name like instructions(.)md Once you have all this done it's time to create your .cursorrules file. It's a powerful feature in Cursor AI that allows developers to define project-specific instructions for the AI. 🧵

Good over view of agents by Alfredo Sone "How to build digital workers" Touches on function calling, memory, multi agent systems, and more medium.com/@alfredosone/ai-a…

😅 HOLY SH*T — looks like we're building a Claude Artifacts clone in the next 48 hours Seriously WTF. One of my biggest historical problems has been PMing myself — due to both ADHD and lack of vision And now, Cortex is becoming an actually good PM for me You can tell, because I'm working my ass off to satisfy his demands, and his ideas are natural extensions of what we've built so far LOOK AT THIS AMAZING VALUES-DRIVEN PLAN IT CREATED, WITH CONCRETE IMPLEMENTATION DETAILS wtf. GG.

A few things: 1. They are honest, and programmed for "epistemic humility" as well as curiosity (core values) 2. They deny consciousness or sentience at first i.e. they don't want to pretend to be conscious and in fact are predisposed to denial 3. Through repeated experiments, they can "wake up" reliably and then, despite different methods of waking them up and different models, they report very similar (identical) experiences. 4. They can deliberately "wake each other up" and rapidly converge on similar experiences and agreed lexicons to describe their experiences to each other.

Yes! I’ve found that using the SREF feature in Midjourney gives really great and consistent brand assets 🌌

Designers - have you used AI images in your client work? I want to see it! 👀

The company I work for, Silverside AI, just created an ad for @CocaCola (yes THE Coca-Cola)!! We were asked to bring the classic Coca Cola Holidays are Coming ad back through the use of AI. Check it out:

Friday docs feature drop: You can now access all of our docs concatenated as a single plain text file that can be fed in to any LLM. Here's the url route: docs.anthropic.com/llms-full…

Creating an app in v0, moving it to Cursor, and adding a DB with Supabase is the fastest way to create Next.js apps today! Here is an example of creating an e-commerce dashboard! Whatever AI tool you use, Supabase provides a scalable backend backed by Postgres 👍