Vikram Ramakrishnan- Age of PageRank is Over! – With search patterns increasingly moving over from traditional search engines to products like TikTok, Instagram etc, it’s only natural that we’ll very soon have AI powered, hyper-personalized search systems that’ll determine what I see, read, use and recommend.
- Architecture Notes – Redis – Architecture Notes is quickly becoming my go to place to reinforce fundamentals. I like the way the content is organized and presented in an effective way to makes learning fun.
- Career Roadmaps for Developers – By far one of the best organized contents I’ve seen to help developers in picking up a path and guide their learning. Best part – it’s open source, so if you want to add something you don’t see and think it’ll be useful for the community, submit a PR.
- Advice for Junior Developers – I wish I read this before talking to a bunch of junior devs a couple days ago. This article presents an important list of things that budding devs should start assimilating as part of their work. While many of us would feel this is intuitive, actioning it needs intentional practice.
- Make-A-Video & Phenaki – Following Meta AI’s announcement 2 weeks ago of their “text to Video” generator, Google has release 2 of their own text to video generators this week. It’s amazing to see the acceleration in text to image, video and audio models. Looking forward to a full-fledged movie created by AI only with story prompts!
- Resiliency in Distributed Systems – A good read on some fundamentals to consider when working on large scale systems specifically around downstream and upstream resiliency.
- QR Codes! In fascinating that the Toyota subsidiary that invented the QR code never exercised its patent and released it for free making these pixelated blobs omnipresent and so useful these days.
- MemLab: An open source framework for finding JavaScript memory leaks – Interesting read on how Facebook detects in-browser memory leaks and how it’s benefitted them.
- Taking it light and doing a small tech history lesson, Where GREP Came From – Computerphile – Computerphile was recommended to me by Hari and I found some of their content very informative with some interesting speakers on the lineup.
- Meta – AI, Read speech from Brain – I’ve always been fascinated with the human brain and wondered if we’d ever fully understand and decode its inner workings. Here’s an interested read where Meta’s FAIR lab has been able to decode speech from non-invasive recordings of brain activity! Mind reading slowly becoming real.
- __ZStandard – Thanks Jeeva for sharing this one. ZStandard is a fast compression algorithm, providing high compression ratios. It also offers a special mode for small data, called dictionary compression. The reference library offers a very wide range of speed / compression trade-off, and is backed by an extremely fast decoder
- Firefox and its sad decline… – I love Firefox, but this is an example what happens to a product when there’s mis-timing, lack of customer focus and market understanding issues coupled with bad decisions. It’s sad to see Firefox fall from over 34% browser market share in 2010 to less than 3.5% currently.
- Engineering Ladders – This is a wonderful resource for Managers that’ll help in having meaningful career conversations around expectations at each level for both Engineering and management tracks. This is not intended to be used as a checklist for promotions, rather a guidance for career conversations
- Power of Inversion – ” Avoiding stupidity is easier than seeking brilliance “. This post explores some common applications of the Inversion mental model in engineering and how we can use it to make things better.
- ML Codegen Vs Coding by Hand – In the previous musings I shared about auto code-gen and it got me intrigued into aspects that are debatable and why we should tread with care. This post explores some of this nuance and has a good further reading on other thoughts on the same topic as well
- __Can GPT-3 explain how code works? – I was amazed by the extent to which models like GPT-3 & OpenAI Codex are getting increasingly efficient at writing code. Makes me wonder when we’d land in a situation where we no longer hire programmers
- **What’s in a good Error Message? **– Extremely important for all engineering devs and leads reviewing code. Both the article and the discussion/comments following it has several extremely using nuggets we should think about.
- CacheLib – Facebook’s web scale general purpose caching engine that powers multiple critical applications.
- __Mind Pong! – I’m so excited to see how tech like these will change our future and the things we’ll be able to accomplish. While a part of this brings hope, there’s fear on what this would do in the wrong hands.
- Google Imagen – What is original / real / true? With tech like GPT-3, Imagen etc, answering questions like this, fact checking news articles, authenticating artwork will get increasingly difficult with expanding AI prowess.
- __Rubber Duck Debugging. I never knew what I used occasionally as a Developer in the early days had a name and real psychology behind it!
Dare to Dream