Monday, May 15, 2023

Error converting content: marked is not a function

location:: Home - Avalon, Boston 
victory-hour:: no๐Ÿ˜ด
pomodoro-blocks:: 4 
wake-up:: 
day:: Monday
icon:: ๐Ÿ“…๐ŸŒŸ๐ŸŒŸ๐ŸŒŸ
  - **08:09 AM, day. Morning Pages**
- Remember: Give it to Self
- โšก๏ธ๐Ÿ““ ๐Ÿ“– ๐Ÿ’ช โ˜ฏ๏ธ โ„๏ธ ๐Ÿง˜ 6 - 8 AM: practice/victoryroutine๐ŸŒž
- #nowPlaying๐ŸŽต #radio/lane8
- ๐Ÿ™† 10 mins #Workout
- ๐Ÿง˜ 5 mins #meditation/invitation20
- ๐Ÿ“– 50 mins #practice/Reading๐Ÿ“–
- #Thoughts
- Americas- reformer
- I came to Mumbai at 14
- Charlatans- @Brittain Laid
- Comment and opine on energies and phenomenon. Ignore people and products #goodone
- 2 profound articles! Inundated with fullness of perspective #tech/ai
- [Generative Turn for Tech Strategy](https://dmccreary.medium.com/the-generative-turn-for-tech-strategy-502390aafb76) by @Dan McCreary
- โ€œSimilarity search is an embarrassingly-parallel operation and can be speeded up by many orders of magnitudeย [using low-cost FPGAs](https://info.tigergraph.com/hubfs/Graph%20+%20AI%20World/Graph-AI-FPGA.pptx.pdf). Donโ€™t let anyone on your staff tell you they canโ€™t find the 100 most similar items from a collection of 10 million in under 50 milliseconds.โ€
  - The inherently recursive patterns in LangChain use a well-traveledย [content enrichment pattern](https://www.enterpriseintegrationpatterns.com/patterns/messaging/DataEnricher.html). But the combination of prompt enrichment with embeddings and large language models was a huge step in the right direction of building agile agent frameworks. It reflects many of the same processes the human neocortex does in problem-solving: reflecting, planning, and filling knowledge gaps using inference.
  - #bookAcquired Books/Timeless Way of Building
- #projectidea platform for integrating PKMs
- Connect various vector db like logseq and create knowledge base
- Platform so this can be done for other pkms
- Platform because product will come form community rather than me
- #Thoughts
  - when itโ€™s hard to build a centralize product; letโ€™s build a platform. Itโ€™s the open source way of building the product hmm.
  - For #RenovateAI App - we expose an API
  - Chatbot interface?
- ๐Ÿ‘ง โ˜•๏ธ 7:30 - 8 AM: #saie/bootime - beautiful boo time
- ๐ŸŒŸ ๐Ÿ–ฅ 9:30 - 11:30 AM: P/AI Expertise. Pumped. Another 2 hours of solid #practice/coding๐Ÿง‘โ€๐Ÿ’ป ๐Ÿ…๐Ÿ…๐Ÿ…๐Ÿ… Day 45
  collapsed:: true
- A cue for building a all encompassing, all data sources, personal agent
  - Ask - "what movie did I like in 2023" - it scans through disparate vector dbs - social, PKM etc. This distinction can allow agent question to be filtered.
  - ![image.png](../assets/image_1684165614106_0.png)
- #TrialAndError.AI - Building an App Template
  collapsed:: true
- Expo, Redux - opinionated tech stack. Start with iOS. Clone to be production ready.
- Bottom Nav. Customizable names for tabs.
- Login. Email, Phone, Google, Apple.
- Apple subscription. Modal screen. - data via config or api
- Use Setting. See plan. Delete user.
- Invite User - contacts.
- Event tracking module - firebase
- App rating module
- Push Notifications module
- Error message toast
- Superset this with- AI App Template
- 1:30 - 3:30 PM: P/AI Expertise Session 2. Let's go. Phew Tired.
- #Logseq plugin development took the juice out. It's the chaos. The unknown. The startup cost you  know. Will get there P/Logseq QA. I have a nice react based HMR based setup ready. So now folder selection, index, and QA UI. Something then I could use from web :))
- ๐Ÿ› ๐Ÿ˜ด 4 PM: #powernap20๐Ÿ˜ด solid 15 mins
- ๐Ÿ–ฅ ๐Ÿ› ๏ธ 4:30 - 6:30 - Build
- โŒ ๐Ÿ’ช 6 - 7 PM: #practice/2WW๐Ÿ’ช
- ๐ŸŒŸ Every Day Kaizen
- practice/breathing๐Ÿ‘ƒ
- practice/accent
  - Accent/d
  - Mud Puddle
- ๐ŸŒŸ๏ธ Daily 5
  collapsed:: true
- What is a tiny step I could make to achieve my goal?
- AI expertise
- Renovate ai - task for the day
- Apply jobs
- Read
- Accent practice