Rami's Readings #125 - Join My Submission to Canada's AI Leadership Consultation 🇨🇦
The latest on AI, LLMs, Canada's AI Leadership Consultation, Effective Context Engineering, Sora, Mechanical Watches, Distributed Systems, Modern Sort Algorithms, and more.
Welcome to Rami’s Readings #125 - 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.
👋🏼 Welcome New Subscribers
Hello! A hearty thank you for subscribing to Rami's Readings! There are quite a few new subscribers this week, thanks to recommendations from The AI Ethics Brief, Life in Conversation with Salmi, Product Byte, and The VC Corner. I am thrilled to have you on board! In this newsletter, I curate the best papers, tweets, and articles I have read during the week focusing on LLMs, AI, economics, business, and technology news. You can learn more about me on my website.
📈 Top Recent Editions According to Substack
Dear subscribers,
I’m ramping back up after a well-needed family break and plan to return to my regular weekly publishing schedule. It feels great to be writing this newsletter again. I appreciate all the messages throughout the break.
🇨🇦 Canada’s AI Leadership Call for Consultation
Canada’s newly formed AI Strategy Task Force launched an open call for public submissions to shape the next chapter of Canada’s AI leadership. I have a few ideas for how Canada can leverage open-source small language models and local, private AI inference to advance the safe adoption and commercialization of AI. I’ll be preparing a submission to the task force (due October 31st) and would love to collaborate with other industry experts.
If you’re interested, please reach out to me by replying to this newsletter or messaging on LinkedIn.
P.S. You can also help by sharing this message. I’m looking to collaborate with up to three experts who have deployed SLMs into production use cases using local hardware for inference (ideally in Canada, but I’m open). To ensure we can complete the submission in time, please get in touch by October 10th.
🤖 AI Reads
Effective Context Engineering for AI Agents
Notes: Another fantastic article from Anthropic.
Privacy-First Voice-to-Text with AI Enhancement for macOS
Notes: Open source WhisperKit & MLX project.
NVIDIA / TensorRT-Model-Optimizer
A unified library of state-of-the-art model optimization techniques like quantization, pruning, distillation, speculative decoding, etc. It compresses deep learning models for downstream deployment frameworks like TensorRT-LLM or TensorRT to optimize inference speed.
Notes: It’s nice to see all the techniques in one place even if this project is optimized for NVIDIA.
Hugging Face’s finepdfs Dataset
Notes: So much of the world’s data is trapped in PDFs.
💼 Business Reads
OpenAI Releases Social App for Sharing AI Videos From Sora
Notes: I have more questions than notes.
The Case for Mechanical Watches in a Digital Age
When I wear my mechanical watch, I’m free.
Notes: 100% agree and my own love for the analog continues beyond mechanical watches.
🔀 Other Reads
Beyond OpenMP in C++ & Rust: Taskflow, Rayon, Fork Union 🍴
Processing Strings 109x Faster than Nvidia on H100
Notes: Both of these articles are from Ash Vardanian and worth the technical read for those inclined.
Notes on Distributed Systems for Young Bloods
Notes: Great read on building distributed systems. Despite being written in 2013-2016, it’s still highly relevant.
The Unreasonable Effectiveness of Modern Sort Algorithms
Author’s conclusion and opinion: It’s absolutely possible to beat even the best sort implementations with domain specific knowledge, careful benchmarking and an understanding of CPU micro-architectures.
Notes: I love these deep-dive investigations.
Signing off from Redmond, WA.