Nostr and CAMPING would be AMAZING 🤩 Camping is my favorite 🏕️ 🌙 🔥 View quoted note →

TIL Grok is a bitcoiner 😂 I opened twitter for 5 seconds and found a thread where grok was doing their best to orange-pill some fiat fellows who were actively dismissing any evidence shared their way 🤷‍♀️ Honestly never felt more empathy for an AI lol 🧡

I’m realizing in the same way that I don’t need to understand how to craft a saddle because I know how to drive a car, I don’t need to understand how to write code because I can vibe a tool that solves a real problem for me That said, I’m learning a lot more about writing code in the process than I’ll ever learn about saddlery 🐎

Grateful we have a place like @Mile Zero Marine to host our @757BTC meetups 🫶🏼 Shoutout to @cnix for providing a space we can vibe and this sick MZM coozie 🍺 (well worth the sats! 🧡)

Day 1 of Vibe Coding with shakespeare.diy I read this article

Soapbox

🔧 Understanding Nostr Service Providers (NSPs): The Future of Decentralized AI Services - Soapbox Blog

Discover how Nostr Service Providers (NSPs) are revolutionizing AI services through decentralization. Learn how to choose the right NSP for your Sh...

but I'm still a little fuzzy on what an NSP is. As part of an effort to be more active with vibe coding today (and ultimately learn more coding over time) I spent about $100 using shakespeare.diy to vibe code a website I could use to help me do my job more effectively. I didn't spend too much time on it, just submitted prompts here and there as I found time throughout the work day. I had mixed results, but I'm also pretty inexperienced so that could have been a big factor. I made something cool, but I have yet to deploy it and I've run out of credits. I wish I understood better how it worked, but this is part of the learning process I guess. For example, how is the cost calculated? I learned before I ran out of credits that I didn't have a backend set up and that my data is stored in my browser right now (so my testing data is not secure as I'd like it to be). How does the cost compare to using tools like Claude AI and Cursor? I wish I could see the code at the same time and edit it directly in real time. Maybe something like cursor is a bit better for that. There was also mention in the aricle above about using more affordable NSPs for lower lift tasks but the only option I could easily see was the MKStack NSP (this one is paid per credits used) What makes using an NSP more private/secure than other centralized tools if it's still using models made by Anthropic or OpenAI? Could I save some money by using locally run LLMs on my own hardware? I really liked how I could see the app in real time as I modified it, that was very helpful and cool. I think this is an affordable and very quick way to spin up a tool that solves real problems, but also expensive if I really can do something similar on other centralized solutions for $20/month. if anyone is open to providing feedback, I need some help getting the app deployed and getting a true backend and database hooked up to it so others can test it with me. If anyone has any advice I'm all ears! #asknostr #vibecoding #shakespeareDIY @Derek Ross