Links #6
THE computer science book,a checklist of mental models,emerging LLM architectures and more
Hi, writing this after a long break. Almost one year. Last post was in June 2022. Will try sending these out every month or two from now.
My interests have changed over the year, so has the name of newsletter (if you didn’t notice).
The Log : What every Software engineer should know about real time data’s unifying abstraction
This is a fantastic introduction to the distributed log and all of the benefits it can bring to a project as scaling becomes more complex. The blog delves into the significance of logs in achieving data synchronization, consensus, and data integration across distributed systems. Reading it is like listening to an Architect thinking out loud. Length of the post combined with its density makes it a bit challenging to get through in one reading.
The Computer Science Book
Concise book available as free to read website. It covers some important aspects of CS relevant to software engineering. I found it useful to learn automata theory and Operating systems.
As I quickly lose motivation while reading books like Three easy Pieces or when Jeff Ullman himself is teaching you pumping lemma.
It is written intentionally to cut down on fluff or whatever Author thinks is unimportant. Worth a shot if you learn breadth first.
Checklist of mental Models
Mental models are how we interpret and make sense of the world around us. Akin to Folk wisdom, they’re powerful enough to quickly improve your decisions or change the way you think.
This is a handy checklist to go through them once in a while.
If you are hearing the term “Mental model” for the first time, then this page on Farnam Street blog serves as an excellent starting point.
New Gatekeepers - Benedict Evans
I discovered Benedict Evans recently. His essays are like Stratechery but with a broader view of tech trends. He has an uncanny ability to distill complex topics into accessible narratives for a wider audience.
Every year he curates an extensive presentation which explores macro trends in tech industry. Here is his most recent presentation titled New Gatekeepers. Check out his website for more of his essays.
Emerging LLM architectures
Deploying LLMs in production is hard due to their novelty and distinct nature. Moreover, many enterprise offerings for LLMs are still in beta phase. This post outlines how to leverage existing ecosystem and proposes alternatives at every critical stage while scaling.
That wraps up this edition. Thanks for reading.
Links in this issue
The Log : What every Software engineer should know about real time data’s unifying abstraction
Mental Models - Best way to make intelligent decisions | Farnam Street Blog
Emerging Architectures for LLM Applications | Andreessen Horowitz