Recent content in shrsv on The Learning Loom
I was browsing some Bitbucket API documentation somewhere when I read something like - Atlassian IPO’d with a market capitalization of 4.37 Billion I wanted to make sense of that sentence a bit more than I already did. Obvious - but worth stating for the record: Atlassian builds Bitbucket. Atlassian was publicly listed on NASDAQ on December 10, 2015. The symbol for the company was the…
The Power of Betting in Decision-Making Navigating Uncertainty In real-world decision-making, uncertainty is a constant. Whether it’s a project deadline, a hiring choice, or a career move, information is often incomplete or unclear. Waiting for perfect clarity isn’t always an option—time runs out, resources dwindle, and complexity can overwhelm. Many rely on believing —trusting intuition,…
It’s not every day that I find myself at odds with someone like Jeff Bezos (on a matter of principle, of course). The man who built Amazon into a global phenomenon once said, “Good intentions don’t work. Mechanisms do.” At first glance, it’s a compelling statement—practical, grounded, and appealing to the results-driven crowd in business and economic worlds. But the more I reflect on it, the more…
Some organizations stand the test of time, lasting centuries or even millennia . They don’t chase rapid growth , which often leads to collapse. Think of a company like WeWork , that expanded too fast and fell apart due to mismanagement . In contrast, some of these enduring groups focus on purpose , stability , and community . For example, small family-owned businesses often endure because they…
“Up, down, up, down, up, down, …” What comes next? You know the answer - it is “Up” How did you know it? Your mind inferred some sort of structure, and made a guess I mean - we do not know - what truly will come next. But we are pretty sure - due to the repetition - and we expect it to be “up”. There could be a breakage of pattern, but the fact that we…
In this post, I continue the exploration from the past post , where I was trying to learn a bit more about the human body from the very basics. MacroNutrients, Et Cetera First, we take in food - as solids/liquids. And that food is broken down into different macronutrients (and other components): proteins: body-building nutrients, found in meat, eggs, beans carbohydrates: sugars & starches, found…
I was recently reading Biohacking Lite by Andrej Karpathy, particularly the section called The Four Batteries of Your Body . For the first time in my life, health-related topics are starting to make sense to me because Karpathy explains them from first principles. TLDR This is my TLDR after an hour of studying the topic. Any mistakes or misunderstandings are my own, and I may update this as my…
I’ve always been interested in decisions—two ways, really. I wanted to make better decisions. I wanted to make decisions in a better manner. When it comes to making decisions better, what I care about is conserving energy while still getting a decision done and executed . In my previous post , I explained how vagueness is the enemy of action . In this post, I’ll share a trick to fight…
I wasted some time today—I realized it when I looked back at how the time slipped away. Naturally, I wondered: what went wrong here? My mind churned for a bit, wrestling with the question, until an answer popped up: vague thoughts . See, I’d been chasing correct judgment, aiming for correct thoughts. In that pursuit of getting it right, I ended up making things more complex . And that complexity?…
These are some LLM-assisted exploration notes from the paper HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Models The Multi-Hop Problem in RAGs The idea of “hops” are important in RAG. Consider this example. Question 1: “Who wrote Hamlet?” → The answer (Shakespeare) is in one document.* Question 2: “Which university did the president of…
A Few Ideas About RAG As I keep pouring in efforts to build LiveAPI , I get new ideas on how to make it better from time to time. There is a constant feeling that maybe we should have some sort of Chatbot on top of the structured API information we generate at scale. People say: “Nobody reads documentation these days” “I don’t think anyone will read docs anymore”…
In Smithsonian National Museum of American History - the following device is visible: (src) The device is called the Mark I Perceptron . It was built in 1957. The chief person behind construction of this device was Frank Rosenblatt The device had 3 parts: Stimuli receptor monitor (S) Association machinery (A) Response mechanism (S) This is almost a mechanical equivalent of what Minsky sort of…
Mastery Demands Payment in Life Units Let’s think a bit about the 10,000 hour rule. It says that mastery in a serious discipline demands a minimum focused investment of around 10,000 hours in deliberate practice . Deliberate practice means - finding benchmark behaviors, breaking them down, and performing them consistently to get nearer to perfection. 10,000 hours roughly breaks down like…
A Definition of Leadership These days the word leadership has become loaded with many meanings, explanations, and imagery across various media. But in its essence - leadership simply means to lead . The roots of the word lead emerge from an Old English word laeden which means “to go first, to guide”. Therefore - leadership is closely tied to: Taking action Taking action before others…
People often say, “Focus is about saying No.” But experience tells us it is hard to say “no” to things we want to do. And every “no” that you muster chips away at your limited supply of willpower. So, how can we make focus practical? That is, what does it mean to arrange for practical focus? Replace “no” with “not now.” How? Many people…
Imagine you’re deep in a coding session, cruising through your repository, when bam—you hit an API reference that stops you cold. You’ve got no idea where it lives, and now you’re stuck. Sound familiar? That’s the pain I want to talk about today, and how a tool I’ve been tinkering with—LiveAPI—tackles it head-on. This isn’t just about saving time; it’s about keeping your flow intact and your…
Hi there! I’m Shrijith Venkatrama, founder of Hexmos. Right now, I’m building LiveAPI , a tool designed to make generating API documentation from your code incredibly easy. Over the past few months, we’ve been hard at work refining LiveAPI to be more robust, scalable, and user-friendly. Here’s a look at what we’ve been up to. What We Did in November and December Towards the end of last year,…
Steve Jobs said taste is paramount to business leaders who wish to do well. I think that’s a bit too abstract to put into practice on a day-to-day basis. Instead, I prefer thinking of it this way: “Let’s not put more garbage into the world.” That is, it is a virtue to be critical about one’s own works. When our standards are high, and we refuse to pollute the lives of…
Learning is movement, radical change. It is not about sitting still and absorbing information. It is about moving forward and making things happen.
There are essentially two types of people: those who are successful and those who ask defocusing questions. A defocusing question is something like: “What can I do about this?”. This creates a huge list of busywork for you – which keeps you running around for a long time with very little results to show for it. Instead, try a FOCUSING question: “What is ONE thing if I do,…
Hi there! I’m Shrijith Venkatrama, founder of Hexmos. Right now, I’m building LiveAPI , a tool that makes generating API docs from your code ridiculously easy. In this hands-on tutorial, we’re going to explore the vanishing gradient problem in deep neural networks—a tricky issue that can slow down or even stop learning in early layers. You can run the sample code in a Jupyter Notebook, see…
Hi there! I’m Shrijith Venkatrama, founder of Hexmos. Right now, I’m building LiveAPI , a tool that makes generating API docs from your code ridiculously easy. Tensors are a fundamental concept in machine learning and deep learning. In this tutorial, we will explore tensors in a fun and light-hearted way to gain some familiarity What is a Tensor? A tensor is a multi-dimensional array that…
The Problem As a founder of a startup building from India, I often ask myself what is the situation of Indian startups in the global market. In startup literature across the web, we often hear about the 1 million in USD ARR as a benchmark for a startup to be considered a “serious” company. So I wanted to know, how many Indian startups already possess 1M+ USD in ARR? A Bit of…