Error converting content: marked is not a function
location:: Home - Avalon, Boston victory-hour:: pomodoro-blocks:: wake-up:: day:: icon:: 📅🗓⚡️🌟🌟🌟🌟 - My Daily Mantra: Honor the present moment - **21:08 AM, day. Morning Pages** - Remember: To Be, and To Imagine - lol it’s lights off - 9 AM - 11 : P/Algorithms. Day 85. Having fun thinking and twisting code. 🍅🍅🍅🍅 collapsed:: true - Algo/Water Area ✅ - Wow. So the key is breaking down a problem into linear steps. Obvious but now I #grok it. As I did it myself with 30 mins on intense thinking on - Boom Right on the money. 90 mins. I wrote, hand twisted, the code, linear fucking time, to make it pass. Worth the screenshot. As I do I have the algo somewhat! collapsed:: true -  - Extra 30 mins. I did it. No hints. Fixed the bugs! Thank you. Took 2 hours with distractions. - Condensed to 11 beautiful lines of code. Love #Algorithims - ```python def waterArea(heights): surp = [0 for h in heights] area = 0 for i in range(1, len(heights)): for j in range(0, i): if heights[i] > surp[j]: l = (i-j) -1 h = heights[j] - surp[j] if heights[i] >= heights[j] else heights[i] - surp[j] surp[j] += h area += l*h return area ``` - 11 - 12 PM: #workworkwork - 12 - ... PM: Work and TV - RenovateAI App - Applied for DUNS Number. Weird freaking Apple requirement. Paid $229 to get it in 8 days or less - #thefuq. First expense from business card Mercury Bank. - 2 PM: Break - 2:30 - 5 - tinkering and researching - Wow. So somebody made P/AI Whisper Transcribe collapsed:: true - https://all-in-on-ai.vercel.app/ collapsed:: true - {{tweet https://twitter.com/transitive_bs/status/1605529637936717826?s=46&t=iA-YyBIc6qDa2ecSdKNf2Q}} - Open sourced semantic searching and indexing built on top of Open AI. inspired by this guy collapsed:: true - {{tweet https://twitter.com/rileytomasek/status/1603854647575384067?s=46&t=iA-YyBIc6qDa2ecSdKNf2Q}} - Damn #mofo😡 - “Transcriptions are generated using Whisper and then embedded using the latest @openai text-embedding-ada-002 model. The vectors are then stored in a @pinecone vector database. A user's query is embedded and then used to find similar vectors in the database.” - The "Ask" answer uses text-davinci-003 to answer the question given the search results, with instructions not to make stuff up. - Another one podcast summary - #lol We've been thinking about making this for 10 years, but it wasn't till @OpenAI came out with Whisper & GPT3 that it was cheap enough. We used @BananaDev_ to make transcripts with @OpenAI Whisper and wowzers (!) @BananaDev_ was so much better, faster, and cheaper than everyone els, I ended up becoming an investor. - Interesting - “Figuring out good prompts was harder than I expected, I suspect this will get easier with GPT4. We used @humanloop to quickly prototype a bunch of prompts to see what would generate the best results. Prompt design moved to 10x faster thanks to @humanloop.” -  - How can I use AI powered search in in real estate context.. ooo ?? #P/Renovate AI - “What about taking it one step further - service that lets you upload a YouTube playlist -> get a semantic search engine back?” - Looks like these guys Kailua Labs are on it -“We're building that @kailualabs 😁 we already have a simple API that accepts RSS feeds & builds you a semantic search engine, and video is coming soon” - The next #Tech thing I picked from the podcast transcription is with Vector Index for search. Hmmm. V interesting. Saying that without knowing anything. But apparently open ai model cna create embedding in vector index that cna then make searches semanti - {{tweet https://twitter.com/hwchase17/status/1605594269795901449?s=61&t=o7TfbXLFwfa8QHWzrPc0Bg}} - 5pm: #tv/myfirstmillion podcast with @Billy McFarland. Dude just got out of jail post Fyre festival crime. He did black card lol - reminds me of jugaadu guy from Mumbai who like luxury, party and spending money. - **21:08 PM Lights Off** - early, fasting today -