(This is a now page, and if you have your own site, you should make one too.) (Last updated: 2026-05-20) Right now I’m an intern member of technical staff at 8090 Solutions, I’ve been accepted to present a poster at IEEE qCCL 2026, so I’m currently revising my final paper and creating my poster, On the side I’m trying to become a more skilled researcher, whatever that means.
Hello! My name is Amitav! A good summary of today is that I did some work that then was invalidated, so I’m feeling a bit deflated. I did quite a bit of work for 8090, but wasted quite a bit of time on setting up some infrastructure that was then nullified when my boss and I decided that it would make more sense for the system I’m working with to work with lighter-weight tools that would allow me to iterate faster as the system I’m working on will likely never see absurd levels of traffic. That’s good for future Amitav, however it is somewhat saddening for current Amitav as quite a bit of work will be redone, but it is what it is I suppose. In addition to that, today I read this article which was interesting (it was shared with me yesterday and I did read it yesterday, but did not share it on this blog). There are a couple of action items there that I will not act on right now due to greater priorities, but that I defer instead to the 28th (the day after final paper submissions are due for qCCL). In particular, I lack mathematical maturity, which is bad and must be solved. Another interesting thing I read today was this slideshow by my boss from a few years ago (not recommended to me by him, found it on my own), though I lack a lot of prerequisites in distributed systems and so it’s mostly gibberish to me. One thing I’d like to do eventually is learn more about distributed systems. Today, I also listened to the podcast by David Senra on Claude Shannon. Overall it was interesting, I think my biggest takeaway is that you shouldn’t take things too seriously. Now, back to Amitav, the tasks I set out for myself yesterday are as follows: ...
Via slashdot. The news of the layoffs themselves aren’t that interesting, but the most interesting part was the news about them recording their employees to train their models. There’s more about this here, overall I think while I wouldn’t like to work somewhere where I’d be recorded for the entire day, for the purposes of replacing me no less, it is in interesting approach to gathering data. Overall though it feels that Meta is losing the AI race. Despite Llama being one of the first open-weight LLMs to gain a lot of mindshare, they’ve sort of stagnated. ...
Hello, I am Amitav! This is actually my daily update to be first posted on my personal website, amitav.net, before being syndicated to substack (see this). The biggest reason for this is that I have an admin portal now on my website, which gives me a snazzy interface with which to write my posts. You can see this for yourself here, though I will warn you that this is password restricted so that I don’t get random folks posting stuff on my website. Anyways, back to the task at hand, my daily update(s)! ...
Hello! This is my first admin portal post, my previous posts have been uploaded manually using scp. The admin portal was created by my openclaw and allows me to write posts in a nice interface and is password protected.
The disconnect between Canada’s intellectual strength and its lack of private innovation fascinates and concerns me. Canada has produced pioneers like Geoffrey Hinton and Yoshua Bengio, the Godfathers of AI, and innovations like the Canadarm and CANDU reactor. Yet few of these breakthroughs translate to jobs and prosperity here at home. Research is essential, but you can’t build an economy on research alone: research proposes value, while business creates value. ...
I’ve been thinking a lot lately about how one teaches the concepts of functions that operate at higher-dimensions, like a function that takes two inputs and returns two outputs. This is hard, because most of our… Note: This post is a work in progress.
Hello! This page is on the rules of engagement on this website. “I want to get more content from you, but I’m worried I’ll forget about your existence” Has this ever happened to you? We have a revolutionary new technology to ease your woes: RSS. I use an RSS reader for most of my browsing, and I would highly recommend you do as well! The one I use is Bubo Reader, which I use because it’s extremely simple, and can be deployed on your own server. You can also deploy it on Netlify, as seen in this demo. ...
What the hell is a matrix, anyways? When I say matrix, you probably don’t have a great idea of what a matrix actually is, and that’s what I aim to solve. What the hell is a vector, really? This section draws heavily on the textbook Linear Algebra Done Right by Sheldon Axler In the previous issue I didn’t go that deeply into the mathematical definition behind a vector. Sadly, before we can truly understand what a matrix is, we must first understand what a vector is. ...
Hello! This is part 3 in a TBD part series on creating an LLM from scratch! You can see part 2, creating lists, here, and part 4, creating matrices, here. What is a vector? For our purposes, a vector is an ordered list of numbers. Here’s an example of a 2-dimensional vector: [5, 7] And here’s an example of a 5-dimensional vector: [3.5, 4, 2, 1.2, 943.89] We use multi-dimensional vectors to represent things like words and sentences and paragraphs, because the hope is that each vector will allow us to encode multiple independent aspects of a token at once. At a high-level, we’re guessing that the vector will begin to associate different numbers with different attributes: maybe we want the first number to roughly correlate to how dog-like the word is, the second number to correlate to royalty, the third number to correlate with age, etc. For more information, see this paper on word representations in vector space. ...