Rami's Readings #91 - Announcing Apéro & Intellect in Seattle and Montréal!
Announcing Apéro & Intellect #11 and #12, and the latest on AI, LLMs, Qwen, New Game Theory, Corporate Bonds, and more.
Welcome to Rami’s Readings #91 - a weekly digest of interesting articles, papers, videos, and X threads from my various sources across the Internet. Expect a list of reads covering AI, technology, business, culture, fashion, travel, and more. Learn about what I do at ramisayar.com/about.
Apéro & Intellect Is Coming to Seattle and Montréal!
Apéro & Intellect is a curated series of intimate gatherings discussing Artificial Intelligence designed to stimulate both your mind and your palate. I am pleased to announce Apéro & Intellect #11 will take place in 🏙 Seattle on Sunday, December 15th, 2024 at 12pm. First time in Seattle!
Apéro & Intellect #12 will take place in ⚜️ Montréal on Saturday, December 28th, 2024 at 11:30am. It will be fun!
#10 in NYC is sold out.
Happy Belated Thanksgiving! Dear subscribers, I’m deeply grateful for your support and engagement in this newsletter. Y’all are the best!
🤖 AI Reads
LazyGraphRAG: Setting a New Standard for Quality and Cost
Notes: Great paper! Use smaller models and defer LLM inference to reduce cost.
Alibaba Releases an ‘Open’ Challenger to OpenAI’s o1 Reasoning Model
Notes: Alibaba’s Qwen family of models are becoming more impressive. QwQ-32B-Preview performs almost as well as o1 on a number of benchmarks. You can download and run it locally with Ollama.
Former Android Leaders Are Building an ‘Operating System for AI Agents’
Notes: A massive round raised at an almost unicorn status valuation. Interesting to see what they are working on.
andrewyng / aisuite: Simple, Unified Interface to Multiple Generative AI Providers
Notes: Many frameworks already exist for unifying GenAI interfaces. This library is interesting, mostly because it was published by Andrew Ng (and/or his team). It is worth keeping an eye on it for that reason alone.
💼 Business Reads
The New Game Theory
Notes: Interesting proposal for an alternative to Nash. If you care about climate change or economic models, you might be interested in reading this paper.
This paper argues that game theory designers should design structures in which there is no incentive to deviate or cheat, and that barring that, players should form pacts or accords with other players to pursue a “second best” equilibrium, which in the end is better in terms of utility both for the individual and for the collective than if they pursue their pure self interest alone.
Rakuten Plans Return to Dollar Bond Market With Hybrid Notes
Notes: Corporate bonds are back! To be fair, they never went anywhere, but let’s do half a billion at these interest rates! 🤑
What the "Superforecasters" Predict for Major Events in 2025
Notes: It’s that time of the year!
Signing off from Redmond.