Ashwin's blog

Stand-up comedy: Rung de

Sun, 26 Apr 2026 00:00:00 +0000

Laugh tracks

So it’s been a little less than a year since I started trying to do comedy. And let me share what I think the milestones in a stand-up comic’s career look like.

There’s one track that goes:

Your first open mic
Your first 3-minute set
Your first 5-minute set or your “tight five”.
Your first 10-minute set
Your first 20-minute set
Your first 40-minute set
Your first special/hour-long show.

And on a separate track you have:

Everything takes more time than expected

Thu, 26 Feb 2026 00:00:00 +0000

A friend suggested that I allow users to login to hunchle.com without a password. This seemed fairly simple to implement, so I thought I’d give it a shot. I was allowing users to login with emails and passwords already, so removing a password was all that might be needed, right? And I had AI to help! Turns out, wrong! First, I had to streamline the new experience so that it worked for both existing users (who had passwords set) and new ones (who might never have a password). Adding this in meant I’d need a streamlined experience, so then I had to spend a few minutes researching how other websites did it. AI takes care of the coding, but needs a big helping hand with all of this. Taste is what I provide.

Miles vs Cash-back Calculator

Wed, 11 Feb 2026 00:00:00 +0000

Miles vs Cash-back Calculator

Miles per dollar Cash-back rate (%) Investment return (%) Devaluation rate (%) Years Redemption rates — CPP (comma-separated)

CPP	Devaluation	Initial Miles Value	Breakeven Year	Miles @ End	Cash @ End

CAGR Calculator

Thu, 05 Feb 2026 00:00:00 +0000

CAGR Calculator

Enter any 3 values to calculate the 4th. CAGR = Compound Annual Growth Rate.

Vibes

Thu, 05 Feb 2026 00:00:00 +0000

Tools and pages I’ve built when inspiration struck.

One-offs

Cash vs Points - Miles vs cash-back calculator with breakeven analysis (11 Feb 2026)
Lichess - Visualize rating progression over number of games played (07 Jan 2026)
Clean Logs - Tool to remove file:line prefixes from log output (13 Nov 2025)
Values - Personal values and life philosophy (05 Nov 2025)
Lennon or McCartney - Beatles preference analyzer based on composition (28 Apr 2025)
Phones - Device Speedometer benchmark scores comparison chart (03 Apr 2025)
Yearly - Year planner with monthly tasks and progress tracking (09 Sep 2024; not updated)
Benchmarks - Personal exercise/fitness benchmarks (03 Jul 2024; not updated)

Smorgasbord: Explainers, Finance, Curiosities, Neural Net Course (In Progress)

An ode to my achamma: What does legacy mean?

Sat, 31 Jan 2026 00:00:00 +0000

I looked up what the latest AIs had said about my father, my mother and the male ancestors on her side of the family. It’s incredible how many things individual people can do. As a constant presence in “my” home for the first 18 years, whatever they did or achieved was just lost in the mundane swings of time that constitute daily life. It’s good to have a retrospective and reflect once in a while. As the Foo Fighters sang, “There goes my hero… he’s ordinary.”

Optimizing Tack On

Sat, 17 Jan 2026 00:00:00 +0000

With AI advancing rapidly, one of the most asked questions these days is, “Are software engineers still needed?”. I have some answers! I love optimizations and performance engineering, and decided to dig into my codebase (with Claude Code) to find possible improvements in this domain. If you’ve worked with these tools, then you know that for an inquizitive person, there is nothing better as with AI, you get to learn and improve at a rapid pace!

I need more context

Fri, 16 Jan 2026 00:00:00 +0000

Click the buttons to rewrite the article in the style of different authors

Happy New Year: Friends Forever?

Wed, 31 Dec 2025 21:38:37 +0000

For as long as I can remember, I’ve found I’m happiest when surrounded by friends. From my early childhood, all the way to adulthood, I spent a lot of time with friends and placed a huge amount of importance on friendships. Throughout my school years, I mostly picked my subjects and activities based on what my friends did - whether it was quizzing, or choosing computer science as the added elective in the 11th-grade (or initially preferring “engineering drawing”) and of course, the sports I played (well, you do need friends to play with after all). This carried on to college, where, while everyone was worried about their grades and career, my plan - just like before - was simply to do what my friends were doing. People were aiming for jobs with top companies while I honestly was aiming to land in the same city as my friends (that didn’t happen, but I made some new ones). And then came independent adult life, where I decided to move out west partly because none of my friends were in the same city as I was anymore. Honestly, I hit “hard relate” when I saw this kid’s answer here.

F1 AI benchmarks

Sun, 21 Dec 2025 00:00:00 +0000

Any researchers who want the prompts and/or links to the chats can contact me =)

I went through the Formula 1 EOY quiz on Mr V’s Garage’s YouTube channel and decided to put some AIs through the test as well. For the LLMs, I gave each of them just 1 shot at the questions. These all use the free versions of the LLMs. Here is how I fared, and how the AIs fared (x indicates incorrect, v indicates correct, p indicates partially correct):

Now

Sun, 21 Dec 2025 00:00:00 +0000

This is my now page. It tells you what I’m currently focused on.

What I’m working on

Improving Hunchle.
In-progress: Writing explainer articles.
On-and-off: Working on an AI course.
Backlog: Working on a language-learning course.

What I’m learning

AI (LLMs from the ground up).

What I’m reading

Tracked on StoryGraph. I prefer this to GoodReads as this is actively developed, and a small business. As a bonus, it’s woman-owned!

Where I am

Western Hemisphere

Are we in a bubble?

Thu, 04 Dec 2025 00:00:00 +0000

Intro/Disclaimer: Trade at your own risk!

Just like women think “I’m special” and “This time it’ll be different” when it comes to relationships, men think the same way when it comes to stock trading.

Warren Buffett’s advice has remained the same for years: Most people should simply dump their money into an index fund that tracks the market and forget about it rather than actively trade and lose money. I too have recommended that same approach in one of my most-read posts, but given my actions, I think I need to add another oft-repeated piece of advice: “do as I say and not as I do.”

YouTube videos

Tue, 02 Dec 2025 00:00:00 +0000

12/02/2025

Jensen’s talk at the Stanford Graduate School of Business (link).

My takeaways

I find Jensen very interesting. He has SRK-like mannerisms, and probably a similar level of charisma (the latest silky, gray-haired Jensen, the Jensen from the mid-2000s, maybe not). He speaks very well, by which I mean he says mostly profound things with very little fluff and is able to talk and deliver his views without making anyone feeling worse. He gives off a huge “sweet talker” vibe (charisma), which is normally something I’ve not seen outside scammers, and to be clear I don’t think he’s anything close to being a scammer.
“Do what you are uniquely positioned to do” (similar to thoughts I had, and also what Satoro Iwata had said). “The only question is, is this important work? And if we didn’t do it, would it have happened without us?” (23:38). “Your job is to make a unique contribution … to do something that nobod else in the world would do or can do” (at 44:50).
Organizations should be structured to enable everyone to do their best work (get out of the way, remove obstacles) “(My job is) to create the conditions by which you can do your life’s work” (twic, at 34:30 and AGAIN! at 43:33).
The CEO (& leaders) should be visible (to employees). “Leaders have to be seen” (27:42). “(if the stock price has dropped by 80%) The most important thing I have to do as the CEO is this: to come and face you, explain it (even if you don’t have a great answer)” (28:35)
The CEO should not rely on secrecy. Information should be shared publicly. “(sarcastically) The knowledge, the information of a CEO is supposedly so valuable, so secretive, you can only share it with two other people or three […] I don’t believe in a culture, an environment where the information you possess is the reason why you have power. I would like all of us to contribute to the company and our position in the company should have something to do with our ability to reason through complicated things, lead other people to achieve greatness, inspire, empower other people…” (33:10)
Regulations should be done by specialists in the field (53:05). Interesting thought. I agree and wonder why this level of nuance is not there in most governments (and for that matter most discussions) around the world. Why not have a young tech advisor or tech-enabler to allow governments to digitize and democratize access to information?

Other notes

He also looks back on life to decide what to do, which (to me) is similar to Bezos’ regret minimization framework. I didn’t understand his point about tokens floating in space (51:10). Quotes indicate I paraphrased something in the talk. They aren’t exact quotations as you may expect, so I have this note here.

I want my Chicken Shawarma: How I cut booking costs by 10x

Thu, 20 Nov 2025 00:00:00 +0000

I like playing sports. I created an app to notify me when courts are available.

General design

I had built this a few months ago and was able to get it running from my laptop. I don’t have my notes from back then.

Adding notifications

What I wanted now was to get it running on a remote server and also to give it the capability to provide notifications. For notifications, I settled on pushbullet though WhatsApp would have been ideal. Seems like more products may be needed in this space to make notifications easier to get. How hard is the WhatsApp API to integrate with? I don’t know, but maybe good to look at for a future project. Twilio is a big player in this space.

Models for thinking

Wed, 12 Nov 2025 00:00:00 +0000

None of these are particularly revolutionary, but it helps to have them listed so you can use them actively to your benefit (rather than hoping these tools fall into your hand).

Line on a spectrum

Many things that seem black-or-white are simply just different lines on a spectrum. Think of the spectrum as a horizontal line, and the “line” as a point (or a vertical line that you wouldn’t cross). For example, for some, any lie is too far. While others may be okay with white lies (“I’m busy today” when you aren’t) and others may be fine with outright lying too (“I know for a fact this product does not harm you” - even when you know it does).

A Mallu in America

Thu, 06 Nov 2025 00:00:00 +0000

Intro

Disclaimer: There are some generalizations below, and so all the disclaimers for generalizations apply. Some of those being: these are inferences being made on a set of observations so the inferences may be completely wrong, and by definition a generalization means this does not apply to all individuals in that society.

The irony of a Malayalee working in America may not be obvious to many. A Malayalee is someone who hails from the state of Kerala where the native tongue is Malayalam. Kerala is home to the world’s first democratically-elected communist government. America, of course, is a capitalistic society.

Pros(e) and concerts

Tue, 28 Oct 2025 00:00:00 +0000

I don’t think concerts are worth going to. Over the years I’ve attended concerts by Muse, Iron Maiden, John Mayer, and others. But, over the past few years, I’ve increasingly felt that the experience of attending a concert isn’t worth the price I’d pay. I’d rather spend time trying out escape rooms, joining sports leagues or if we want to stick to live performances, then watching theatre and stand-up comedy, where tickets are still reasonably priced.

What I've been up to

Fri, 17 Oct 2025 00:00:00 +0000

A laundry list

Started doing stand-up comedy. You can read more here.
The first two weeks after I quit, I did a lot of chores - both for the household and myself.
Started reading about company and business valuations. Now I understand P/E ratios, the PEG model, how to use discounted cash flows to estimate a fair stock price & compare it with the implied P/E ratio and how to determine the health of a business by comparing debt to revenue. I’ve also talked with friends about looking into market cap as a multiple of revenue and using the price-to-sales ratio when a company is not profitable.
Some weeks I did nothing except relax by playing video games or surfing and scrolling.

A Malayalee's guide to Kerala

Mon, 29 Sep 2025 00:00:00 +0000

Kerala

As an NRI, your dollar goes a lot farther in India. On trips back to India, my sister usually finds great resorts for us to enjoy. I’ve since realized many friends within India travel to Kerala as their go-to quick vacation destination, and so, I thought it’d be useful to provide a Malayalee’s guide to vacationing in Kerala. And if at least one person in your party speaks Malayalam, it makes things much easier (to communicate with the locals). If you don’t know Malayalam, English is your best bet (and then Hindi).

EB-2/EB-3 for Indians

Tue, 02 Sep 2025 00:00:00 +0000

The green card blues

If you’re an Indian in the US (or thinking about moving to the US), you likely know the problems with immigrating to the US. By virtue of being born in the most populous country on Earth, the employment-based green card route is effectively closed to Indians.

However, there isn’t a clear way to know how long the wait may be. And if you don’t know how long, then you can’t easily plan what you’d like to do. So, I wrote a script to estimate when your priority date may be current (i.e. when you are able to move forward in the green card process). Ping me directly if you want the predicted date for your priority date.

On Software Design

Mon, 01 Sep 2025 00:00:00 +0000

For a lot of my career, I never really understood software design. It was always presented as something more associated with SDE2s, or higher and not so much with SDE1s. I had no idea what to make of it or what to do with that information. The first few design documents I wrote seemed inane. There was something to be done, and we all knew what we wanted to do, but it looked like I had to formalize it in some way before I could move forward. The document always had a structure, always following a similar pattern, that made this exercise seem even more tedious. The structure usually went like this: Motivation, Problem, Requirements, Chosen proposal, Considered but declined proposal, Appendix. A glossary and “not in scope” section could also be excluded. For each proposal, we’d need to list out the pros and cons.

EB-5 for Indians

Wed, 06 Aug 2025 12:23:34 +0000

DISCLAIMER: None of the below is financial advice or recommendations for anyone beyond myself. Please do your due diligence (and don’t sue me, thanks!)!

What

EB-5 is a program where you invest some money (800K$ minimum) and in exchange may be eligible for a green card.

One of the key issues with the EB-5 visa program is that it is currently backlogged for Indian citizens. However, some specific investment categories (targeted employment areas or TEAs) are not affected by this backlog. It’s essential to ensure that any investments are directed into those categories to avoid delays.

Thoughts on superintelligence

Fri, 01 Aug 2025 14:35:22 +0000

Zuckerberg has written a post on “personal superintelligence” here. Most of it is just fluff - for example, “superintelligence” is never defined*, so it looks more like a PR piece. However, there are a couple of lines worth commenting on.

If trends continue, then you’d expect people to spend less time in productivity software, and more time creating and connecting.

Not sure what “productivity software” refers to here, but as a business that relies on creators, it makes sense that Meta would want people spending more time creating and engaging with their software… err.. I mean “connecting with others”. Sincerely, I do believe Meta’s products facilitate some connection, but my full thoughts on this are complex and better saved for a future post. This seems out of place with the rest of the article (which is about “empowering” people) but is in line with what makes sense for Meta as a business.

Chancing my arm at stand-up comedy

Mon, 14 Jul 2025 00:00:00 +0000

My love for stand-up

Click to expand; feel free to skip

I've often thought of myself as someone who loved making people laugh. I'd played that role at school, though I'm not sure others remember me that way. In college too, I was delighted when others were laughing - whether with me or, occasionally, at me.

I’d been watching stand-up since around 2007, when Russell Peters was all the rage. An Indian-origin stand-up comic was unheard of, and he quickly built up a fanbase in Dubai. He seemed to put in front of people things that were uncomfortable to say, but fun to laugh at, and at the time, that was refreshing. It was also okay, as long as it was confined to the stand-up stage. Thanks to the internet, I quickly followed up with videos of George Carlin, Mitch Hedberg, Eddie Izzard, Louis CK, Jerry Seinfeld and many others.

Hunchle: Generating questions

Tue, 10 Jun 2025 00:00:00 +0000

Time for a blog post detailing the difficulty with creating a fun-to-play trivia app (www.hunchle.com)

Generating questions is hard

A good quiz app needs good questions. That should be obvious, and your #1 priority, but when I initially built Hunchle, I decided to focus on other features (adding AI hints, streaks, share results, and thinking of whether I should include ads: I didn’t). What makes a good question? As someone who has dabbled in quizzing through high school and college, I was told, “A good question is one that you can work out. Not one that you either know or you don’t”. For example, “What is the name of the elf in the Fellowship of the Ring?” is a horrible question. There is no way you can guess the answer if you don’t know it. A much better formulation of the same question would be, “Legolas is this type of being, which shares its name with the title of a movie starring Will Ferrell”. Here, even if you don’t know the word “Elf” from The Lord of the Rings context, you may be able to guess at it from the movie clue.

How I used AI in Hunchle

Wed, 28 May 2025 00:00:00 +0000

I used AI in a big way to build Hunchle. I figured it would be useful to write this post to share my experience and to document what coding with AI looks like today.

There were two major ways I used LLMs in Hunchle, a trivia app:

1. Generating questions

In an ideal world, I’d generate 30,000 questions by myself, custom-crafted the way I like quiz questions. However, that was more of a time investment than I’d have liked. So, I decided to use AI (ChatGPT, Gemini, and Grok mostly) to generate some questions for me.

Democratizing AI

Fri, 25 Apr 2025 00:00:00 +0000

If you’ve been following along, I’ve been using LLMs for a while, as detailed here and here. I learned a lot about the world of LLMs and I’ve tried to capture that here. Recently, motivated by Thorsten Ball’s excellent “How to build an agent” guide, I decided to jump in and get my feet wet building AI agents with AI (LLMs).

Hello, nice to meet you!

Like any good beginning, I must include introductions.

Why LinkedIn is the best social network (for me)

Fri, 18 Apr 2025 00:00:00 +0000

At the outset, this may seem an absurd conclusion. To make my point, I’ll need to go into a little bit about how social networks came to be. It won’t take long.

Back in the mid-2000s, we had a bunch of social networks pop up. Hi5, Orkut, and of course, Facebook got a foothold on Dubai’s “interwebz”. Possibly due to its simplicity or due to the exclusiveness stemming from its invite-only system, Facebook soon came to dominate. Importantly, at least in the beginning, Facebook had a simple guarantee: you post, and it would - with a 100% guarantee - land on your friends’ walls (as feeds were called back then). This meant everything that I wanted to say would get blasted to my friends’ screens and all the cool parties my friends were at would dangle irresistibly out of reach on my feed.

Marriage Penalty Calculator

Sat, 21 Dec 2024 00:00:00 +0000

2025 Marriage Penalty / Bonus Heatmap

Approximate 2025 U.S. Federal Marriage Penalty / Bonus

Simplified model: wages only, standard deduction, and a stylized childless EITC-like credit to produce penalties at low, equal incomes (for 2025 brackets). Positive values are penalties (more tax when married), negative values are bonuses. All numbers approximate, not tax advice. The 2017 Tax Cuts and Jobs Act (TCJA), now effectively locked in through 2025 by later legislation, made almost all the joint brackets exactly double the single brackets and doubled the standard deduction, sharply reducing bracket‑driven marriage penalties for middle and upper‑middle incomes.

Sleep Activity Timeline

Sat, 21 Dec 2024 00:00:00 +0000

Continuous Activity Actogram - Sleep Analysis

Continuous Activity Actogram

Visualization of computer activity blocks over a continuous timeline. 🟢 Green = Active/Awake, 🔴 Red = Sleep, 🟡 Yellow = Untagged gaps (click to tag)

Sleep Duration Analysis

Sat, 21 Dec 2024 00:00:00 +0000

Sleep/Awake Duration Analysis

Sleep Duration

Mean

Std Dev

Awake Duration

Mean

Std Dev

Sleep Duration Distribution

Awake Duration Distribution

A guide to finances: securing the foundation

Sun, 15 Dec 2024 14:30:14 +0000

Intro

Around 7 years ago I wrote “A guide to finances” which some of my friends found helpful. Now, everyone is likely at a different point in their lives, so I thought I’d write the follow-up.

General overview

Write down what your ideal life looks like and calculate how much that would cost per year. Let’s call that THE_GOAL.
Once you have THE_GOAL, divide that number by 0.04 to get how much you’d need to be completely financially independent. If you’re risk-averse, divide by smaller number (0.0375, 0.035 or 0.03) instead.
Once you have that amount, invest it into the market with 80% of it going to stocks, and 20% going to bonds (you can use 85-15 as well).

Chat Statistics

Sun, 24 Nov 2024 00:00:00 +0000

Chat Statistics

Chat Statistics (Anonymized)

Messages per Hour

Characters per Hour

NAND game

Sun, 17 Nov 2024 00:00:00 +0000

What is it

I just finished playing NAND game which gets you to build a computer starting from simple relay circuits. As you build components, you get to reuse them. Over time, this means you have an arsenal of components that let you eventually build a computer. I think this is similar to a course Nand2Tetris that I’d started years ago but never finished. I skipped the software lessons, but completed all the hardware ones.

How I use AI (part deux)

Mon, 11 Nov 2024 00:00:00 +0000

Adding to my list of what I’ve used AI (read as “LLMs”) for. For my previous post, click here.

ChatGPT was quite helpful answering questions about the areas I was visiting in Europe. For example, I was confused by all the buildings and names, and sought help in the form of the question: “What was the Domus Flavia part of?”. ChatGPT answered that it was Domitian’s palace, which among other things included a basilica. That confused me even more, as evident in my follow-up: “Is that the same basilica as the basilica of Constantine?”. The answer was no.

Cancun

Wed, 18 Sep 2024 00:00:00 +0000

Our trip to Cancun

Day 1

Landed on Friday afternoon. Took the ADO bus from the airport. It was like 8$ per person to get to the city. Withdrew ~2000$ in pesos from an ATM at the airport. Don’t choose to convert when the ATM asks. Check online for which ATMs are good to use (in terms of exchange rate).

We stayed in the city, near where the ADO bus drops you off, near the intersection of Avenue Uxmal and Avenue Tulum. I’d recommend not staying in the hotel zone unless you are going there to get the all-inclusive experience and simply want to relax in a hotel all day. Staying in the city is better to explore the “real Mexico” (IMO). After settling in to the Airbnb, we walked to a small restaurant on Avenue Bonampak and grabbed a Michelada (highly recommend!). There were many restaurants here open late at night, like Botamero Tropical (now closed), where we got some free shots in exchange for writing a review online. There were many other restaurants on the street open late as well. The food was generally excellent at all places we tried.

How I use AI

Tue, 10 Sep 2024 00:00:00 +0000

Brief list of what I’ve used AI (read as “LLMs”) for.

What I’ve used AI for

Laziness

I could have googled things in this section and figured my way around it with a little effort.

Writing code to use features that I know exist, but am not proficient at:

I asked it to mock static methods using Mockito. This was a feature introduced in the newer versions of Mockito, so I am not as familiar with mocking statics off the top of my head. The syntax is also somewhat weird, so asking the LLM to do it works well.
I asked it to cherry pick a file from another commit.
I asked it to show me a way to see files in a git stash.
I asked it for a command to copy the last git message. Answer: git log -1 --pretty=%B | pbcopy
Asked it to split a commit into 2 commits. See below.
Asked it how to stash only unstaged changes. Answer: git stash --keep-index

Working with cumbersome CLI commands

I asked it to write a find command and pass in the arguments to sed as the arguments to pass to find are never obvious to me. Specifically, whether the path should come before the --name option or not. I also find (heh) that sending output from find to sed is really hard to get right. Do I get it to work with xargs or pipe the arguments to sed or use the exec option within find? Well, with the LLM I just had to ask what I wanted it to do and the command it gave me worked perfectly (specifying platform of course, because sed behaves differently with in-place replace in files on the Mac v/s Linux).

Sex and gender

Thu, 01 Aug 2024 20:40:41 +0000

Disclaimer 1: This is my understanding, if you spot something wrong, please assume best intentions and let me know, thanks!
Disclaimer 2: All uses of the word male/men/female/women here are subject to the caveats I’ve tried to detail below.

I had to write this post as the topic of sex and gender (in sports) has popped up yet again this Olympics :)

Firstly, sex and gender are not the same. A person’s sex can be male or female, and so can a person’s gender, and that is where the confusion may arise. Let’s dive in.

The art of watching sports

Wed, 26 Jun 2024 01:50:42 +0000

This is a post about watching sports, but it is written not only for those of you that watch sports or are sport-curious (those that think that they may want to watch sports but aren’t sure). This is also a post for anyone who likes to watch movies, or shows, or read books. Lastly, this is also a post for those of you who hope that your spouse may share the couch as you plop down to watch the highlights for the most recent sporting spectacle you’ve gotten into.

Heroes

Tue, 18 Jun 2024 23:35:17 +0000

“Don’t meet your heroes,” I’ve heard. “People are imperfect” I counter.

Here are a few people I look up to, with the reasons I look up to them stated. They may have flaws, but the flaws are not why I look up to them¹. I admire them mainly for the good that they do, and for choosing – in what seems like a batshit crazy move to many - not to prioritize profit².

Cluster headaches

Tue, 13 Feb 2024 19:13:00 +0000

Cluster headaches

I’ve always thought it would be a better use of time to do research on conditions rather than simply pray. In one case, after doctors had near given up on figuring out what was causing a family friend’s son’s high fever, said family friend read up and suggested that his son may have dengue to the doctors. Turns out he did - doctors had not tested for dengue because it wasn’t dengue season.

2023 review

Sun, 31 Dec 2023 00:00:00 +0000

Written after the fact, but here are a collection of my monthly reviews for 2023:

July
June
May
April
March
February
January

As evidenced by the short list, I couldn’t keep up after July

A covid chronicle

Sun, 06 Aug 2023 00:00:00 +0000

The initial signs

Unfortunately, on the last day of a trip to Alaska, I started feeling a throat infection coming on. This was, however, not to be a simple throat infection. A few hours after I’d gotten back home (a day later) a fever of 102.3 had built up.

A citizen scientist

A day later, I decided this might be a good opportunity to chronicle the progression of the disease, and collect some actual data. I started recording my temperature, and the medicines I took. We’ve had a spike in citizen journalism with the advent of the internet, it would be nice to also have a spike in the number of citizen scientists! Here’s a quick graph of what the five days battling with the fever looked like. By day 6, I was fever-free.

Riding the STP

Thu, 20 Jul 2023 00:00:00 +0000

The S T P

Back in June, I heard one of my friends talking about doing the STP with a friend. “Oh, can I join?” I asked. “Yeah, of course, man!” came the reply and in a second I had signed myself up for a 200+ mile bike ride with little idea of how hard 200+ miles would be. As I read up on the internet, I grew more and more terrified, sure none of this would go well for me. Online anonymous commenters were very pessimistic and asked me to be cautious. After a 40-mile ride, I thought they were pretty much right.

The sorry state of software UI/UX

Mon, 17 Apr 2023 00:00:00 +0000

Can you tell the time?

Totally normal clock face

Nope, you aren’t looking at a programmer’s attempt at being funny. This was the official clock app on Android.

This was actually an attempt at improving an earlier iteration, which for some godforsaken reason stuffed the hours from 13 to 24 into the clock face displaying 1-12:

Why settle for one clock face when you can have two?

Inexplicably, 00 is included in the inner clock face, meaning the inner clock face actually goes 00 -> 13 -> 14 -> 15 (and so on)! I’m perplexed this was released to customers. And why would you even want to update a “clock” app? Surely, it can’t be improved anymore… or can it?

The decentralized web

Wed, 22 Mar 2023 00:00:00 +0000

Over the last couple of years there has been a lot of talk about decentralization of the web. I seem to think we already had a decentralized web - back in the early 00s. Before the rise of centralization, everyone had their own little corner of the web. Blogs/spaces/forums were common and even if you didn’t own the domain name, no one was trying* to run a business off of content you provided (Matmice, anyone?). A “private” place on the web, that people owned, that allowed and encouraged everyone to put themselves out there and share their personalities online. This soon changed with the rise of social networks like Orkut, Hi5, MySpace and Facebook, which meant people no longer owned the spaces where they were sharing themselves, leading to eventual centralization and monopolies preventing competition**.

The case for the curmudgeon

Sun, 19 Feb 2023 00:00:00 +0000

The case for the curmudgeon

Ever notice that one guy on the team always complaining about stuff? That’s a curmudgeon!
While today there are calls for more positivity in all spheres of life, I’m going to make a case for why having a curmudgeon on your team might actually be a good thing.

Firstly, a curmudgeon ends up creating a safe space for others on your team to raise concerns. If a curmudgeon’s constant complaining is tolerated, other employees may feel more comfortable sharing any concerns they have, bringing transparency and openness in communication to the workplace. Handled well, this can lend some hard evidence to support the oft-heard phrase “Bring your entire self to work” (with the unsaid second half: “without worrying about it”). Psychological safety based on research by Google was shown to be the number one factor impacting the effectiveness of a team.

Software engineering blogs worth following

Wed, 18 Jan 2023 00:00:00 +0000

Here’s a list of some cool blogs I think are worth a read:

Software engineers

Approximately in the order I’d recommend them in Julia Evans
Marc Brooker
Joel Spolsky
Steve Yegge
Cindy Sridharan
Vaidehi Joshi - not really a blog, but the basecs and baseds series.
Dan Luu

Unranked (haven’t gotten to yet):

Carlos Arguelles (haven’t gotten to it yet)
Brandon Rhodes
explog.in
evalapply.org
the.scapegoat.dev
Manuel’s Obsidian
Lily Mara’s talks
Lily Mara’s blogroll
Mara’s blog
near.blog links
Architecture Notes
Mahdi Yusuf
Franco Fernando’s newsletter
Chris Munns talks

Will add more as I find more worth adding here.

Writing grips

Mon, 16 Jan 2023 00:00:00 +0000

I was unaware that the different writing grips were named and catalogued! I found a list here. When I was young, I believe I used the dynamic quadropod grip and then in the seventh grade noticed that almost everyone was using the dynamic tripod grip. So, I learned how to write with the tripod grip as well.

Today, I use a mixture of both, with a bias towards the quadropod.

Hard work catalog

Sat, 24 Dec 2022 00:00:00 +0000

Ongoing list of people talking

about hard work, practice and preparation

Rahul Dravid

at ~1:30 here.
Also at 3:30 about being prepared.
At 5:02 about practising in the nets (email to Pietersen).
At 7:43 “You can’t always control the results, but you can certainly control your preparation.”

Kapil Dev

at ~4:45 here

Sunil Chhetri

at ~5:00 here.
At 7:47 as well.

Conan O’Brien

at 4:04 here “If you work really hard and you’re kind, amazing things will happen.”

Novak Djokovic

Some predictions

Sun, 20 Nov 2022 00:00:00 +0000

Ongoing list of predictions

Format: [Date], [chance of prediction coming true in my estimation]: [time period]

2022-11-20, 60%: Apple/Google will make a moderation decision on behalf of Twitter by banning it from the app/play store.
2023-07-03, 80%: Twitter’s “rate limit” will be rolled back to allow more views to unverified accounts (currently at 1000) by the end of August.
2023-07-03, 80%: Twitter’s “no access unless logged in” change will be rolled back by the end of August.

Some thoughts

Mon, 31 Oct 2022 00:00:00 +0000

In a stark departure from my usual posts, here is some utterly unscientific nonsense that I mull over from time to time:

You can convince people to believe you are great, if others simply keep repeating that you are. What’s even more surprising, is that you can even do this without the help of others. Simply praise yourself all day, everyday, and take credit for accomplishments that have involved great help from others. It’s even possible if you had nothing to do with the thing in question, and only in some cases is this fraud. However, this is morally wrong, without question (in my book). Which brings me to my next point.
If you wish to be moral, you will get beaten out by someone who doesn’t. Ideally, we would want to stick to our morals all the time. However, being moral is effectively placing a restriction on yourself (like a tax), and one which in most cases leaves little to no benefit other than the upholding of morality. The thin line between morality and legality - sometimes even wandering into legality - is where many “successful” (see previous point) people operate.
If one wishes, they can easily manipulate others. In fact, this is essentially what Dale Carnegie’s book How to win friends and influence people is about. What is more incredible, to me, is that you can even manipulate yourself! I once tried an experiment on myself, where I repeatedly told myself something I didn’t think to be true (that I thought had a < 5% chance of being true). After 2 weeks, I ended the experiment expecting to go back to normal. However, I couldn’t snap out of believing what I used to believe. Unrelated to my next point.
I believe there are 10x engineers, or even 100x engineers. In other fields of individual pursuit - like sports - there are individuals who are clearly more skilled/capable/talented than others, and this is a non-controversial opinion. As of 2022, I do not see why this would not apply to engineering. Which brings me to:
One person can - and has - changed the world. This has happened multiple times in history, and will continue to happen again. For examples, Newton, who revolutionized optics, calculus, physics and other fields, Galois who transformed mathematics before dying at 20 and Ramanujan, whose work I do not understand. Understanding? Ah,
Assuming human intellect is likely on a bell curve (similar to genetic traits like height), it must be really frustrating for people much smarter than me to explain stuff to me. I can see this sometimes because so many concepts are so elementary to them that while their explanations flow, I flounder like a fish on land.
Hard work can, and does, help. I assume “intellect” lies on a bell curve, but that only serves as a potential or some kind of “max limit” you could reach. This range lies much higher than where most people are, and that is because hard work determines how much of that potential you hit. Now, does the capacity for hard work lie on a bell curve?
There is nothing wrong with saying “I don’t know” in response to a question, and that is the right answer for most people for most questions
Social media optimizes for engagement, but when social media companies celebrate daily active users, it is a bit like celebrating giving sugar to a baby (or even an adult). Of course people will love never-ending short form video that supplies unlimited dopamine
Empathy can go a long way. Would never have thought it ten years ago, but empathy can change a person’s mind more effectively than anything else.
Many men are immature (or more comfortable being immature?) compared to women, and this might stem from the power disparity in society (which may stem from the power disparity in physical strength).
Creating the best form of the argument you are responding to, before responding to it, is a great way to have a discussion rather than an argument.

Tech interviews: beating the game

Thu, 10 Feb 2022 00:00:00 +0000

Disclaimer/Foreword: The aim of writing this series of posts is to allow our modern interview processes to be viewed through the lens of history (chronicling the process) and science (identifying problems and inferring improvements from process changes). This is part two in the series, which will focus on how to maximize your chances given the current state of the system. The rest of this post assumes you’ve read part 1, and are interested in interviewing for tech/big tech companies.

Tech interviews: an origin story

Fri, 21 Jan 2022 00:00:00 +0000

Disclaimer/Foreword: The aim of writing this series of posts is to allow our modern interview processes to be viewed through the lens of history (chronicling the process) and science (identifying problems and inferring improvements from process changes).
This series will contain 3 parts:
the first one (this post) describes how interviews got to be the way they are today,
the second (read here) describes how you can maximize your chance of success at these interviews,
and the third is about improving the process and fixing some of the issues plaguing tech interviews.

100 days of chess

Tue, 12 Oct 2021 00:00:00 +0000

Goal

The goal here is to try to do 100 days of chess study and see how that affects my games.

Plan

Will pull from a list of resources I have. Will add books below as I decide to start tasks:

[book] Logical Chess: Move by Move by Irving Chernev.
[book/video series] My 60 memorable games by Bobby Fischer.

MoveGuesser to “learn” the GM games.

Day-by-day logs

Day 1: 10/10 - First two games/chapters from Logical Chess Move by Move (by Irving Chernev). Both games were in the Giucco Piano.
Day 2: 10/13 - IM Sowmya’s explanation of Fischer’s game against Sherwin https://www.youtube.com/watch?v=x2QGuK2dLsA. Featuring the Sicilian, with white playing d3.

Open AI Codex - coding with AI

Thu, 12 Aug 2021 00:00:00 +0000

I got to use the Open AI Codex API today, as part of the Open AI codex challenge. The challenge consisted of 5 problems, meant to be solved as quickly as possible, with the Codex available for help.

Clicking the “Codex it” button would get Codex to “parse” the problem statement, the comments in code and the method signature. Following that, it would write code that solved the problem. For the problems in the contest, the generated code either solved the problem completely or required only minor changes (<= 3 lines).

Reflections on my career

Thu, 27 May 2021 00:00:00 +0000

Having worked for a little over six and a half years as a professional software engineer, now may be a good time to reflect on what I’ve learned and how I’ve grown.

Company 1

In my first role, I primarily worked on implementing bug fixes and small features. While I feel bug fixes are not well considered by the industry, I really enjoyed working on them. They work as great on-boarding tools, allowing developers to come in and simply implement the fix or can also be great tools for learning the code-base. You can only fix a bug when you really understand what is causing it (if not, something will fail sooner or later).

April review

Fri, 30 Apr 2021 00:00:00 +0000

After basically taking all of March “off” (only did Week 1, and lost my records for that week), hoping for a more productive April. Mostly porting March’s goals over.

Habit goals

10 minutes of walking per day. [Same as Feb]. Third month in a row.
2 times a week running (for varying times). [Slight change from Feb]. Tracking two 5Ks a week by week 3.
[Paused] 3 chess puzzles in a row every day.
[Paused start] 1 chess blitz game (3+2) per day. Play & analyze.
3 sets of push-ups every other day. [Same as Feb] Weighted, 3x11 with 2:00 rest between sets. Starting out non-weighted, to ramp up.
3 sets of squats every other day. [Same as Feb] Weighted, 3x15 with 2:00 rest between sets. Starting out non-weighted, to ramp up.
[Secret goal]. [Continuing from Feb]. Targeting 1:1. Setting a low bar so it can be cleared.

Non-habit targets

Complete Designing Data-Intensive Applications (DDIA) by 03/10. Mid-chapter 7 as of 03/01. [To be updated]
Complete Martin Kleppmann’s lecture studies.
Complete FreeCodeCamp’s courses on APIs & microservices.
Write 3 blog articles (can be private).

And now,

Science content on YouTube

Sun, 18 Apr 2021 00:00:00 +0000

Some of my favourite science content on YouTube

PBS Eons
Covers life on earth, including some great videos about early humans.

Veritasium
Pretty cool science channel, where he explains different science phenomena. Based on his philosophy (from his PhD work) that showing people misconceptions and then correcting them is what gets them to learn. Showing the “truth” directly leaves them with the feeling that their preconceived (often incorrect) notions were right. He details this in this video

A guide to code reviews - reviewed!

Wed, 24 Mar 2021 00:00:00 +0000

Over my career as a software developer, I’ve changed how I approach code reviews. Simply put, I think of code in more practical terms and less as an “art”. Concepts of beauty, and perfection are best left to be incrementally achieved, and should always be considered with the end goal in mind (delivering value to the customer). Below, I’ve included my post from 3 years ago, with edits.

This here is a collection of things I’ve learned about reviewing code in the few years that I’ve worked as a software developer. ~~As these are my thoughts, you should take them with at least a pinch of salt instead of at face value, as with most subjective things in life.~~ If you’re new to the role, you may find these helpful.

February review

Sun, 28 Feb 2021 00:00:00 +0000

Started the February “review” a bit early this time, stated my goals, and tracked them more regularly (aiming for every week). I think this gives me a good chance to pivot on goals as necessary.

Habit goals

22 minutes of reading every day. [Drop from Jan] Down from 30 minutes in January, which was a little too much for me. Again, counting only time spent reading books, not articles. Modified to read 1 chapter (of DDIA) per day on Feb 13. Modified to read 1 chapter every 2 days (on Feb 22).
10 minutes of walking per day. Decided not to change this one; added a running goal instead.
2 times a week running (for varying times). [New] Aim is to do 10 mins first week (continuing from January), 15 the second, 22 the third and 25 the fourth.
5 chess puzzles in a row every day. [Same as Jan]. Dropped as a goal for the last week, can still do in free time (on Feb 22).
3 sets of push-ups every other day. [Change from Jan] Weighted, 3x11 with 2:00 rest between sets.
3 sets of squats every other day. [Change from Jan] Weighted, 3x15 with 2:00 rest between sets.
[Secret goal]. Will be revealed later. Upped from 2:0 to 3:1 (on Feb 22).
1 design article per day. [New] Added on Feb 10. Reduced to 2 per week (on Feb 22).

Non-habit targets

500 correct chessCom puzzles for the month. Total of 2744. [SUCCEEDED]
- Week 1: At 2603
- Week 2: At 2770
- Week 3: At 2999
[Added end of week 1] Finish reading & paste notes for Designing Data-Intensive Applications [FAILED]

And now,

January review

Sun, 31 Jan 2021 00:00:00 +0000

After seeing others (Susan Fowler, Julia Evans, etc.) write year-in-review posts, I thought I should give it a go too! However, I don’t trust my discipline to last all year. So, I’ve decided to post monthly reviews.

For January, I set the following goals, all framed in a ‘habit’ style.

30 minutes of reading every day. Reading was a fond childhood habit, which I have replaced with oodles of pointless Redditing, YouTubing and Twittering. As a kid, I’d carry my books along on visits my family made (serving the purpose of a comfort blanket, while also engaging me I guess). Time to get that back on track. I forced myself to only count time spent reading books, not articles.
10 minutes of walking per day. This is some indicator of how unfit I had become at the end of 2020, given that till 2019 I could easily run a 5k in under 28 minutes with little effort.
5 chess puzzles in a row every day. Intended to improve my skills at chess. Based on something I read on KhanAcademy (I think) stating that streaks of correct answers indicate mastery better than %-based metrics.
3 sets of push-ups every other day.
3 sets of squats every other day. For both this and the push-up goal, I didn’t really set a fixed amount of reps per set. This may be something I change going forward.
Watch 2 tech videos every week day. This goal was abandoned after 4 days. I need to fit ’learning’ into my schedule in a different way.

And now,

A little history of philosophy by Nigel Warburton

Thu, 28 Jan 2021 00:00:00 +0000

Overall rating: 9/10 Recommended if: you want to start reading about philosophy Level: Easy, intro-level.

Summary

In short (heh), a great introduction, covering a lot of philosophers. I liked it better than whatever I’ve read of The Story of Philosophy. Sometimes I wonder if Nigel accidentally misrepresented/misfocused what philosophers stood for to ensure transitions between chapters were smooth (for eg:- I’ve seen Locke contrasted with Hobbes in the past, but could draw no such contrast here). The transitions were really smooth though and I am glad I was exposed to many interesting (and some useless) ideas. Gives one a good base to start thinking from.

Same Interview

Sun, 29 Nov 2020 00:00:00 +0000

Fernet

O4fJSlQDKUFguV0mksfMOIVOLD6j9SPc8Fck_SXxCV4=

gAAAAABjhw0NmBgemUC-LjPhCneuiRP3T22ng_unjV1yF4aTsmA0RsgRVRy_FrhgeMDYhwZ7JN85BTErCgiCUa-U2PUPuVujtDy1kfNL7N00JpgSi2uJoPScgyx_msKdL4m_jrKWCgtxzhcFCftKUzApr5s1W4_AgT20TGoNB4W-ARte19EpScGhktRIDkrH-RByzqOIQnfPvp1tolW92ukYuMm7RURdq8FtwDbFv0PqN4UrZ-7qZVzwysR384GrZD2PC-_3UO2HMytXX7agGJCJ_1FeN_7rwb6mvZlNfwDJu7AU0Wy-VQOYcuAYn9ocSEg4GG2Q739Ha27h9U9t5DJPa-5byfWX9FGEIxOZ6dCOCgyf10LE03m3D4Harih4mATAp8V7ChxPwULu38qyFRiAeMdFbq4NqtmE6FMEmPCttxiU9hp8VnzvJ4sa87eetCJIJgHuZ-N_ynzaYsCg0VPX–mmLhUKs3a7ZrdUxd4ga8h3N7VWXS7fPeC3L88zLftJ23rHG658-YoycnVncAGl7F9xB_NGiBuSn-fBB9YUcLxwdLmHWRnOyZH5nE0yjVO4GeoVCbAmVo-mOg5on1IJxacelhw_WYlKGJyBnqWM6FsOpG2RALCOTV7u1xQcwjP4cnGsAq5BrYq2NeHzG9uaYsPK89m0n4PhsRljQOa7XuYpq7Dcfsn5YwFqQk_UJu_rZAX-H_0U_4PToA6nuM8vrzNF-fZ2wTqr-iM4h46iuzTrgP2VBvFqBx4GeeL20Y_pUMUleBs6PggrrmxSYggOFSREXOc-S6q7lRCXeFFU9QJLsT-pKq4Hr7UtJTdLn45pWg8IM6oLlFZ4virMiP9Sl9otBsYefNNZvFqluyr1IufKQwOhfT_IkzsF-d43jx7EioqlpYQMrClInyO9DeB3a50W9hRJfhIa8yY2CM1GqF74OX2o8_Y6Otens5j0yeQzFXKd2WVpGvSDN4aZiTj0zLjZgGWlS7P74o3X3GmWgdNf1F24nz04rThmgKbza7zIwcW3EhYWjtYJoBDbFcG_ZBdT6WQmNEDj1l3EYsXw5SZrY886WFWxXJbjRvL20prokVCQcCatT-KyJJI8Qelh_e_NJeqUOoQw9zpSbaNeJodJAZBGtX1sTepcquk4hwQKlS5Erz3NdUZrS2B6aOJw9bPVJDMI_D5zvqxsaBsS_gZ7WfN0mq_kSEBTn5v4pcBULeti_tnKXcjqyb6_nXkY0cFWg-7hna_AmeHCeT3D_2GOdNgDhuZhToka09Qv17-fNCtRFmfhAaKIeasndk6AFPfdFx-nIXGzJsleS96ZCIjVkyWTxulRSCSTti2VHv2EOkxS5kYQJYEk2qZ_3sJFt_ge4wFkxd5uZpJucICJhBiMFTita4GtrQwSH3AzaJcXZ_WO8hIXDOKtfYQiegK3ifORwCIw2-1uhyUdNiICV7nC6rDzfVbvOlaWujCP2qf8u8WamSXKcJL6Z3TbH2MxjHNS36wuOinAFdnwq3tWXi0PrKN0Xdobj1TBeCT2mBQ84WKhxZQJzgOLeIZMWdMY5h_RoK1CfeF_TJZ8_CmC0ds8dfL9yC5jdicZqSkuSn28sIWxBM3dg_tcBMy0KbbdHxEIzSEDk3_yciztTKNhRzFJjy_t5pWI2zC3eB655RYf8GC8g5a700-tNLZlR3ZKV5EcdwWFuIo3kkU4cCJcsmJSf-eDS6dGy_Xo-hggIbFZuTRYQNjULeORyFYQfnxoOod7S8f2k2FhXB5dxdoMzUjM8HhP4pU41s3C0mf2ZM9qtrPOJbbz-HasaCHlbTHApRDfgnYs-qHDQEBVGq7Wl5bSEUFgW0Y33GUSdZTrcfRz4_psJ680IdhZh4yTUItkge-pGPpSLYwoXuSYgk-51LoCkKfRs5LSYYEMo1ujnzC3ST-9lN9m0w6dDRixw1g28Eq4ZXbZUGF9czFv2vDmYCnlZDH808czK3RcCLCh9e9Pabi_Ad6FAA0I08cwD_54IGpYQCfE7w==

Let's talk about: Health Costs

Mon, 21 Oct 2019 00:00:00 +0000

Drug prices are similar to the points in “Whose Line is it Anyway?” - they’re made up. The price that is shown in the infographic (250$) is purely artificial, considering the drug maker gives a $62.50 rebate, meaning the actual cost to the drug maker is only $187.5.

Image from Kaiser Health News shared under a CC license.

Textual description

What happens is, the company that makes the drug sets a price, let’s say 250$ (this is a “fake” price as you will see). Now, the drug is sold to a wholesaler for a certain price (237.50$ in the figure). The wholesaler then sells it to pharmacies for a profit (240$). Notice how the price dropped from 250 to 237.5? It’s because the 250 price is a lie. The actual price was always 237.50$ and 250$ was chosen to inflate the price and make it seem like the buyer is getting a great “deal”.

Let's talk about: Climate Change

Thu, 10 Oct 2019 21:08:00 +0000

Talk by Chimamanda Ngozi Adichie https://www.youtube..com/watch?v=hg3umXU_qWc

Let's talk about: Climate Change

Thu, 10 Oct 2019 21:08:00 +0000

With all the recent news and coverage that climate change is getting, it is easy for one to get a sense of ennui, to develop a numbness to each new report that paints a grimmer version of the future compared to the previous one. It is easy to feel helpless, as if relying on the ability of a captain on a sinking ship. This post is written as an attempt to try and get you to disavow that notion.

Of flights and men

Thu, 20 Dec 2018 23:22:00 +0000

“Woah!” he chortled.
“I look different, I guess.”
“You look like a writer.”

I hadn’t seen Ashish for quite some time. We went to the same school, but only struck up a friendship towards the end of high school. We lived awfully close (less than a km away from each other), but it was only by the eleventh grade that my parents deemed I could walk a kilometer safely by myself. Not soon after, school ended, and we went our separate ways to college. Now, nine years on, a shaped beard hugged his formerly chubby face.

Let's talk about encryption - again!

Tue, 20 Nov 2018 23:51:00 +0000

This post is part of a multi-part series.

For part 1, click here.

Wolf Einstein

The wolf has realized you’ve foiled his initial plans, and he’s decided to up his game. Now, he knows you encrypt your message with your private key. He decides he will decrypt the message with your public key, which will let him know exactly when you plan to visit. Then, he plans to intercept you on the way to your grandmother’s house - just like the fairytale foretold.

Let's talk about encryption

Mon, 05 Nov 2018 23:51:00 +0000

The better to write you with

Let’s say you are Red Riding Hood, and you want to write a letter to your grandmother telling her you will be visiting. However, the big bad wolf is aware of your plans, and he decides he will also write a letter to your grandmother (pretending to be you) saying you won’t be visiting. If only there was a way you could let your grandmother know your letter was from you and the wolf’s letter was a fake. What if I tell you, little Red Riding Hood, that there is a way you can do this!

Let's talk about encryption - part 2

Thu, 01 Nov 2018 23:51:00 +0000

This post is part of a series. For part 1, click here

The problem

We have two computers (let’s call them the client and the server) that want to communicate securely with each other. This means they want to ensure that the messages go to who they want it to go to (the other computer & not some random imposter-computer) and that the messages do NOT go to anyone else (like some spy-computer trying to steal the message).

A guide to charity

Tue, 18 Sep 2018 19:22:00 +0000

Intro

I’ve always had a few thoughts about charity floating around in my head. This post is a way for me to record some of those thoughts.

Deciding where to give

For me, this has been the most important question when donating any money/time/effort. If you have X dollars to donate, where should you donate this money?

Goal 1: Effective allocation

No one wants to see their money wasted. The primary concerns I had were that the money would not be used appropriately (either the charity is a front for a scam, or just inefficient in the way it utilizes funds). This is solved well enough for my needs by sites like GiveWell, and Giving What We Can. These sites rank charities and give you an easily accessible list of charities that have the highest impact.

Optimizing Outlook

Wed, 11 Apr 2018 19:29:00 +0000

Introduction

They say “necessity is the mother of invention”. It’s incredible the power programming gives you. Maybe if the power-hungry learned how to program they could be constructive rather than destructive. Then again, the recent Facebook scandal makes you wonder.

The problem

Stepping down a few rungs and we’re back to the little world I live in. For the past few days/weeks, I had been frustrated by the lack of mailbox space at work. 2 GB. “But, mail providers give you 15 GB for personal email nowadays!” I hear you say.

A rating system for table tennis

Mon, 12 Mar 2018 17:29:00 +0000

Background

At work, during breaks, most of my team plays table-tennis. This weekend, I decided to spend some time writing up a rating system that I could use to keep track of how good (or bad) I (and everyone else) was. This is a writeup of how that exercise went. I’ve hidden a few lines to act as spoiler (indicated by an arrowhead to the left of the line). If you wish to exercise your mathematical skills, try making a guess and then click the line to reveal how I solved the problem.

The quiz-biz

Wed, 07 Feb 2018 23:17:00 +0000

Intro

For a large part of my life, from when I was about 10 years old till I left college, I used to take part in quizzing competitions. A couple of times, I’ve had people ask me how they could start quizzing, or what preparing for a quiz would entail. This is a question I have some difficulty answering, because I have never really prepared for a quiz. In fact, procrastinating while trying to prepare is pretty much how I learned some of the different tidbits that have helped me win different quizzing competitions. Looking back at quizzes I’ve participated in the US, India and the UAE, I notice that in a manner somewhat evocative of the plot of “Slumdog Millionaire”, most of what I know is just due to “life”. An open mind, an inkling of inquisitiveness and a few fortuitous forays into (frankly) frivolous fields would aid anyone on a quest in the quizzing world. Here, I recount three snippets from life that helped me on the journey.

A guide to code reviews

Mon, 29 Jan 2018 01:34:00 +0000

This here is a collection of things I’ve learned about reviewing code in the few years that I’ve worked as a software developer. As these are my thoughts, you should take them with at least a pinch of salt instead of at face value, as with most subjective things in life.

When I first started, I reviewed code very differently from how I review code now. Despite this being a very important skill, it is not taught in high school or even college. Like with most other engineering skills, you’re tossed into the deep end and left to learn how to swim by yourself.

My weird walk to work

Tue, 10 Oct 2017 19:01:00 +0000

The walk

Intro

I’m sure a few other people do, or try to do this, so I’m not sure how unusual it is, but I thought I’d post about it regardless. I walk to work. Almost everyday. And usually (like a good engineer ;) ), if I do something repeatedly, I try to optimize it. With code, with life. So, now that it’s been a few months of walking to work in Seattle, I think I can present my learnings.

H*cking around

Sat, 23 Sep 2017 22:29:00 +0000

Intro

People think hacking requires a great deal of skill (on the part of the attacker) and/or a great deal of skill (on the part of the defender). This is quite clearly false. There are many simple ways to exploit a system. In this post, I detail some of the times I got access to accounts I shouldn’t have been able to get access to. While these are not the specific ways that a system’s bugs are taken advantage of, in general the principles hold true.

2FA, phishing and U2F

Wed, 06 Sep 2017 23:50:00 +0000

Intro

As part of efforts to improve security, my workplace has issued us Yubikeys. What is a Yubikey? It’s a device you plug into your computer’s USB port and when you touch it for a few seconds, it spits out a string. This string is used for authentication (to identify you). The device has a capacitive sensor to detect touch, just like most phones today (old phones that used a “stylus” use resistive-touch based on resistors). A month or so ago, I saw a video/article stating that 2FA (two-factor authentication) was susceptible to phishing, and there exists a new proposal U2F that could fix that issue. I knew that the Yubikey supports U2F, but I didn’t understand how it solved the problem.

A guide to finances

Mon, 30 Jan 2017 19:22:00 +0000

Intro

I write this article to provide an easy way for people just entering the workforce (many of my friends) to manage their finances. I also think this article would also be a good reference for me, and may prove interesting (in hindsight) in a few years. If you want to spend as little time as possible, skip to the end. I would appreciate comments if you find this useful.
Note: this is not going to be rigorous, but it should make a fair bit of sense. Also, this article is pretty US-centric, so if you live outside the US, this may just be a “fun read” (if you find finance “fun”).

Ashwin helps an Olympic medalist

Wed, 02 Nov 2016 00:21:00 +0000

Part 1

As an Indian watching the Olympics, there is not much to look forward to. The best in our one billion-strong nation don’t make it past the qualification criteria for most sports. And, when they do, they are usually found at the bottom of the standings - forming a solid base from which, every Olympics, we moan the dearth of infrastructure and the deluge of corruption in India.

The list of medals won per capita at the Summer Olympics has India finishing last in Total Medals per Capita, Gold Medals per Capita and second-to-last (woot! woot!) in Weighted Medals per Capita, finishing ahead of that great Olympic nation - Iraq, who were nice enough to only win one bronze medal ever.

Love, Drugs And Rock N' Roll

Mon, 22 Aug 2016 23:41:00 +0000

[Disclaimer: I have no degrees in Psychology, Biology or Anthropology. I would love to receive comments, clarifications or corrections.]

Love

I was never insane except upon occasions when my heart was touched

Edgar Allan Poe

Love - a feeling succinctly defined by the Oxford dictionary as “an intense liking for another person”. An emotion intense enough to form the backdrop of literary and cinematic tragedies - Romeo and Juliet, Joel and Clementine and my personal favourite, Eloisa to Abelard¹. What motivated the extreme actions taken by these characters? What could cause this unbearable wanting to be with another?

Let's talk about: The Olympics (Match Sprint)

Tue, 29 Mar 2016 08:00:00 +0000

Sprint: To race or move at full speed (Dictionary.com)

Given the above definition, it shouldn’t surprise you that the fastest cyclists in a match sprint reach speeds as high as 70 km/h. Watching the event may leave you confused - it looks like they are fighting to lose the race. The sprint shown below looks, at times, like a slow-cycling event:

What’s happening here is that being in second-place, right behind the leader, reduces drag (friction with the air) by quite a large amount - around 30%. Less drag means you cannot just race at full speed from the start. If you did, the cyclist in second-place would need to use less energy and can pull ahead in the final stretch to clinch first place. This adds another element to the race - the element of surprise. The leader has to constantly look behind to spot exactly when the other cyclist is going to “make his move”. Miss that moment, and your race may be done, like Chris Hoy learned here. Theo Bos, the winning cyclist, tried that again, but it didn’t work as well.

Let's talk about: Dinosaurs

Thu, 10 Mar 2016 02:08:00 +0000

Dinosaurs.

Intro - Archosaurs

We begin the journey around 250 million years ago, at the end of the Permian era, marked by the largest mass extinction in Earth’s history. The Permian-Triassic extinction event was catastrophic, claiming around 90 percent of marine life and 70 percent of life on land. Synapsids, animals that dominated the Permian landscape, were all but wiped out. A few trickled through, biding their time to regain their kingdom in a later era, but for now what was important was that life had entered the Triassic era - albeit barely.

The joy of tennis

Thu, 01 Oct 2015 03:08:00 +0000

Intro

Sports - activities where people try to prove they are better than others at an activity they wouldn’t be caught dead doing otherwise. Outside the world of sports, it would be ludicrous to find 11 people running behind a ball, but say you can do it better than any other set of 11 people, and the world will throw millions at you. While they might seem stupid if you try thinking about the higher purpose in the purpose of sports, you would be hard pressed to find someone not tuning in for the big final. Not sure why that is. Maybe it’s the privilege of getting to see two masters battle it out to prove themselves. Or maybe it’s the vicarious wave of victory that sweeps over you when the underdog comes out on top. Or maybe it is the mawkish mess you find yourself reduced to when your team or your player finally wins that coveted trophy. Or maybe it’s something else all together.

Probability and tennis

Sun, 31 May 2015 18:09:00 +0000

The idea

The other day, while surfing Reddit, I came across a few comments that said having a 55% chance of winning a point would give you a 99% chance of winning the game. I’d heard this before, but this time I thought, “Hey! I can simulate that and see if it makes sense.” Seemed a bit unrealistic to me, that 50% would mean dead-even whereas 55% would mean a blowout.

First race

Tue, 12 May 2015 22:09:00 +0000

Ran my first race on May 2nd. A 20 km run around Lake Monona.

Started off too fast - too confident despite not having trained since the end of March. Soon, I was tired enough that I would end up walking not insignificant distances throughout the race. Even with all the walking and the pleasant weather, 20 km was really exhausting. Not sure if it’s a Wisconsin thing, but two pints of beer is what you get after you cross the finish line. Just what you need when you’re dehydrated and exhausted as hell (I’m kidding).

Linear circuits

Sun, 05 Apr 2015 23:40:00 +0000

Link to course
University: Georgia Institute of Technology
Overall rating: 4/5
Recommended if: You want to know what would be taught in your first circuits course in college or are just interested in refreshing what you learned about circuits.
Level: Intro
Good: Covers and explains the basics in an easy to understand manner. Discussion forums were helpful. Lots of numerical problems to ensure you understand a decent amount.
Bad: Some topics are glossed over - like transformers and the wheatstone bridge. The lab is weak and boring, when similar courses have much better labs. Proofs and derivations are sacrificed in lieu of just providing a formula, more so as the course progresses.
Interaction with other students: (4/5) A lot of interaction on the piazza forum, mainly due to a few students who were very active answering questions. An instructor goes through the forum fairly frequently too.
Work assigned: (4/5) A good amount of homework and a quiz for each lesson too. Some homework questions (the “select all that are right” type) seem a little confusing and unnecessary. Though, everything considered, doing the work will help you understand better.
Lectures: (3/5) Slides with an instructor speaking over them. Nothing too fancy, but not too shabby either.
My final score: 100%

Dino 101

Mon, 16 Mar 2015 23:35:32 +0000

Link to course
University: University of Alberta
Overall rating: 4/5
Recommended if: You are mildly interested in dinosaurs, evolution or animals. Also, if you would like to know what paleontologists do.
Level: Intro
Good: A great intro to dinosaurs and paleontology. The instructors take feedback seriously and have changed the course since the first offering.
Bad: Wish it went into more detail and was more challenging. It felt like a course without a lot of college level content. More like a fun documentary. Practical/lab work is limited - you may say understandably so.
Interaction with other students: (3/5) Discussion forums are fairly active and there are some interesting discussions held. Instructor participation is minimal.
Work assigned: (2/5) Very little. Should take you around 10 minutes to get a full score on all assignments. No lab work, no assignments other than a multiple-choice quiz for each lesson.
Lectures: (4/5) Well produced and informative. The artwork during the lectures is great too. Some lectures may be a tad bit too long.
My final score: 99%

Learning how to Learn

Sun, 08 Mar 2015 12:35:32 +0000

Link to course (Now self-paced)
University: UCSD
Overall rating: 2/5
Recommended if: you like self-help.
Level: Easier than an intro
Good: Organized, has a structure.
Bad: Nothing that can’t be learned by browsing the web while procrastinating.
Interaction with other students: (3/5) Fair amount of discussion on the forums.
Work assigned: (1/5) Just some written assignments. Don’t feel like it really helped you learn the content (oddly enough).
Lectures: (2/5) A bit like watching a documentary, except a little more academic. I’d rather get this info from a BBC documentary. The stuff that wasn’t tested (interviews and such) were more interesting.
My final score: Dropped out. Not worth it.

First post

Thu, 05 Mar 2015 19:35:32 +0000

Hi, and welcome to my blog.

Here, I will keep you up to date on the things I am working on. Hope you enjoy your stay.

Mon, 01 Jan 0001 00:00:00 +0000

I write for people younger than me, in the hope they can gain from what I've learned through experience. I believe sharing knowledge is one of the greatest accelerants for societal progress. The printing press was revolutionary precisely because you didn't have to experience things to learn from them - you could simply read about others' experiences. Even works of fiction are great because they allow you to experience distant worlds - worlds that don't exist, worlds separated in time and location. I was not alive during the English Civil War, but thanks to the magic of books, I can place myself as a member of society experiencing that period of upheaval.

In recent times, the internet has supercharged this. There existed a balance: the young had the better capacity to learn, but the old had more material (years of experience) to learn from. The internet is the great equalizer in the battle between neural plasticity and experience. When a seventeen-year-old today can access the accumulated knowledge of centuries, the gap closes fast. As evidence, you can look up how much younger the top ten chess players of today are compared to fifty years ago.

Respect for elders across cultures is a reflection of the advantage that age once brought. In today's world, that advantage is greatly diminished, and society needs to shift respect away from experience and towards the pursuit of knowledge and the truth. This blog, at times, is my contribution to that grand process towards progress.

Mon, 01 Jan 0001 00:00:00 +0000

Benchmarks

Freemotion bike, level 7, 5.78 miles, 20 mins, 78 rpm, 190 cal

Mon, 01 Jan 0001 00:00:00 +0000

Log Cleaner

Paste Logs

Cleaned Output

Mon, 01 Jan 0001 00:00:00 +0000

The “old” internet: https://projects.kwon.nyc/internet-is-fun/ Networking: https://twitter.com/iximiuz/status/1859339483390296354 https://fabiensanglard.net/quake3/network.php

AI: https://www.cs.toronto.edu/~duvenaud/distill_bayes_net/public/

WebRTC: https://twitter.com/arcinston/status/1858603234455126441

DL: https://fleuret.org/francois/lbdl.html https://fleuret.org/dlc/ https://spinningup.openai.com/

s3 https://simonwillison.net/2024/Nov/26/s3-conditional-writes/ https://docs.google.com/document/d/1yzLXSOtzBXyaWHfeVsWsMu4xmOH8rV6QyM5ZAnJZjMQ/edit?tab=t.0#heading=h.yhvnt89pggpj

data structures https://x.com/debasishg/status/1858560338456952998

Observability https://news.ycombinator.com/item?id=39529775 https://www.honeycomb.io/blog/how-are-structured-logs-different-from-events https://x.com/mipsytipsy/status/1858341191932338354 https://www.honeycomb.io/blog/get-infinite-custom-metrics-for-free https://x.com/mipsytipsy/status/1857626227634524541 https://x.com/walktothesun/status/1837657459869339834 https://x.com/mipsytipsy/status/1857599831470125461 https://bsky.app/profile/mipsytipsy.bsky.social/post/3l77h7eyfwe2b https://jeremymorrell.dev/blog/a-practitioners-guide-to-wide-events/ https://bsky.app/profile/jeremymorrell.dev/post/3lasvhflczb2u

Bluesky https://x.com/Werner/status/1859611503537594880 https://newsletter.pragmaticengineer.com/p/bluesky-engineering-culture

Access tokens https://twitter.com/gunnarmorling/status/1860961333409104353

Caching https://x.com/hnasr/status/1861197188371751021

Design/architecture https://thefrugalarchitect.com/

Testing https://til.simonwillison.net/pytest/pytest-recording-vcr

LLM/genai https://elk.zone/techhub.social/@simon@simonwillison.net/113539371054898619 https://elk.zone/techhub.social/@simon@simonwillison.net/113541020246317028 https://blog.dottxt.co/coalescence.html

Databases https://www.scylladb.com/2024/11/25/database-internals-working-with-io/

Formal methods https://cacm.acm.org/research/how-amazon-web-services-uses-formal-methods/ https://www.datadoghq.com/blog/engineering/formal-modeling-and-simulation/

Deterministic simulation testing https://notes.eatonphil.com/2024-08-20-deterministic-simulation-testing.html https://www.warpstream.com/blog/deterministic-simulation-testing-for-our-entire-saas

Distributed counter https://netflixtechblog.com/netflixs-distributed-counter-abstraction-8d0c45eb66b2 https://news.ycombinator.com/item?id=42129097

Serverless https://developers.podcast.go-aws.com/web/podcasts/episode_137/index.html https://docs.aws.amazon.com/lambda/latest/dg/foundation-progmodel.html

Career framework https://dropbox.github.io/dbx-career-framework/

Interactive https://www.boot.dev/tracks/backend cryptopals codecrafters https://rubymonk.com/learning/books/4-ruby-primer-ascent/chapters/18-blocks/lessons/51-new-lesson

https://app.codecrafters.io/ https://www.deep-ml.com/ https://viewsourcecode.org/snaptoken/kilo/index.html

web https://elk.zone/techhub.social/@b0rk_reruns@jvns.ca/113544995299428026

Leadership/non-technical https://x.com/scarletinked/status/1861053227472478317

Papers https://ia802709.us.archive.org/12/items/polya-how-to-solve-it/Polya%20-%20How%20to%20solve%20it.pdf On writing (low prio): https://screencraft.org/wp-content/uploads/2018/03/David-Mamet-Memo.pdf https://arxiv.org/pdf/1409.0003 https://samoburja.com/gft/ (low prio): https://worrydream.com/refs/Nelson_T_1974_-_Computer_Lib,_Dream_Machines.pdf (low prio): https://www.maths.ed.ac.uk/~v1ranick/papers/wigner.pdf https://archive.org/details/zero-to-one-book https://med.emory.edu/departments/neurology/_documents/cousteau_swimming_sharks.pdf

Mon, 01 Jan 0001 00:00:00 +0000

https://en.wikipedia.org/wiki/Ullatan_language https://en.wikipedia.org/wiki/Thachanadan_language https://www.youtube.com/watch?v=kXPc4KogBtQ https://www.youtube.com/watch?v=FEB9y7yqj9o https://www.youtube.com/watch?v=Wai9EaylkRQ https://www.youtube.com/watch?v=RhqGD362nso https://www.youtube.com/watch?v=xp06DP-LcN4 https://en.wikipedia.org/wiki/Vishavan_language https://en.wikipedia.org/wiki/Wayanad_Chetti_language

Mon, 01 Jan 0001 00:00:00 +0000

CoffeeStudy PeanutExperiment Ones Complement JVM working Sound digitization (Other senses) What is a kernel (linux)? collaborative book

Mon, 01 Jan 0001 00:00:00 +0000

Loose motion Bunk Hill station Dickey Doubt Ladies finger Hostel Footpath Creche Tiffin Prepone Banian To check: Brinjal Garage Garbage/Trash Servant/maid Cradle Nipple (pacifier) Increment Bumper Bogey Running shoes

Mon, 01 Jan 0001 00:00:00 +0000

Understanding JPEG

https://gist.github.com/DavidBuchanan314/c05a834c485af83f51c25c81c35974ce https://bsky.app/profile/retr0.id/post/3lbscemto2c2h

Mon, 01 Jan 0001 00:00:00 +0000

“I feel really alone. Whether it was a democratic country or an authoritarian country, all of them acted the same when it came to us. I discovered that my humanity is a subject of discussion. Should I be killed in the street or not be killed? Based on the interests of the United States or Saudi Arabia or other countries. It’s the killing of a dream” - https://youtu.be/xaTKDMYOBOU?list=PLJ8cMiYb3G5eqy5GUl0V3MFUSiMyPF8qp

Mon, 01 Jan 0001 00:00:00 +0000

Falling in love You are the one that I want Summer nights Andar conmigo Quelqu’un ma dit Eres para mi Mi tesoro Head over feet Que veux tu Espacio sideral Ojos as� Rosas Lento Gotitas de amor No soy una de esas Me quiero enamorar Something just like this

In love Limon y sal Chocolate A Dios le pido Me llora el cielo Fast car

Break up songs: Heading towards Slow dancing in a burning room Con qui�n se queda el perro

Mon, 01 Jan 0001 00:00:00 +0000

X & O

fingers in hand

Mon, 01 Jan 0001 00:00:00 +0000

Beatles Song Preference Analyzer

Song Comparison

Selection History

Begin selecting songs to determine your preference

Song A

Listen to the song

Song B

Listen to the song

Round	Song A	Song B	Your Choice

Mon, 01 Jan 0001 00:00:00 +0000

Lichess Rating Graph

Username

Game Mode

Max Games

Mon, 01 Jan 0001 00:00:00 +0000

Device Speedometer scores

Mon, 01 Jan 0001 00:00:00 +0000

Values

Feel free to fork and create your own set of values and/or suggest changes via a pull-request at this link.

Having clear values helps lead you to happiness. Without a clear set of values, which can help you with your priorities, you may be lost and engage in sub-par actions for both yourself & humanity.

Take care of yourself. This doesn't mean "spoil yourself". The line between the two forms a spectrum and you need to be clear where it crosses over into "spoiling yourself" territory.
Try not to harm. At a high level, try not to harm. I say try, because there could be cases where you undertake an action that has some downstream impacts causing harm (like working in tech could cause resources to be constrained in a particular geographic location). This has some interplay with the first point.
Spread the knowledge. If you know something useful (let's say you have understood how to start a business), share it. Don't gatekeep it for personal gain.
Reduce waste. From the "reuse, reduce, recycle" mantra. However, I'm not sure consumerism is all bad as it may lead to growth in the economy, benefiting all. However, holding this value means you'd rather reuse things before buying something new, even if the new thing is cheap.
Push for the truth. Whenever possible and wherever possible, push for the truth. People may make short-term sub-optimal decisions based on lies or what they believe is true due to a poor understanding of things. You should strive to seek the truth & share it. Engage to learn and find what may be true, not to prove a point.
Let live. From "live and let live" & stoicism. Focus on yourself. You can't really force others to do anything, so just do your best, do your part. Others may arrive at different conclusions, at which you can try to see where you differ and explain what drove your conclusions.

Footnote: My list of heroes may shed further light on my implicit values.

Mon, 01 Jan 0001 00:00:00 +0000

Planare - Yearly Planner

Loan Interest Rate Calculator

Calculate the Effective Annual Rate (EAR) for an installment plan.

Lesson 1

Mon, 01 Jan 0001 00:00:00 +0000

Lesson 1 Gradient Descent

Lesson 1: Neural Networks with PyTorch¶

📚 Prerequisites (Your Starting Level)¶

Based on your background:

My returns against VOO

Mon, 01 Jan 0001 00:00:00 +0000

Stock pricer

Mon, 01 Jan 0001 00:00:00 +0000

Stock price estimator

Stock pricer (stock price estimator)

What this tool does (click to expand)

This tool estimates the fair value of a stock based on your assumptions about earnings, required return, and future growth. It projects earnings forward year by year, discounts them back to the present using your required return, and converts that into either a fair P/E ratio and a fair price if the EPS is provided.

Ashwin's blog

Stand-up comedy: Rung de

Laugh tracks

Everything takes more time than expected

Miles vs Cash-back Calculator

Miles vs Cash-back Calculator

CAGR Calculator

CAGR Calculator

Vibes

One-offs

An ode to my achamma: What does legacy mean?

Optimizing Tack On

I need more context

Happy New Year: Friends Forever?

F1 AI benchmarks

Now

What I’m working on

What I’m learning

What I’m reading

Where I am

Are we in a bubble?

YouTube videos

12/02/2025

My takeaways

Other notes

I want my Chicken Shawarma: How I cut booking costs by 10x

General design

Adding notifications

Models for thinking

Line on a spectrum

A Mallu in America

Intro

Pros(e) and concerts

What I've been up to

A laundry list

A Malayalee's guide to Kerala

Kerala

EB-2/EB-3 for Indians

The green card blues

On Software Design

EB-5 for Indians

What

Thoughts on superintelligence

Chancing my arm at stand-up comedy

My love for stand-up

Hunchle: Generating questions

Generating questions is hard

How I used AI in Hunchle

1. Generating questions

Democratizing AI

Hello, nice to meet you!

Why LinkedIn is the best social network (for me)

Marriage Penalty Calculator

Sleep Activity Timeline

Sleep Duration Analysis

Sleep/Awake Duration Analysis

Sleep Duration

Awake Duration

Sleep Duration Distribution

Awake Duration Distribution

A guide to finances: securing the foundation

Intro

General overview

Chat Statistics

Chat Statistics (Anonymized)

Messages per Hour

Characters per Hour

NAND game

What is it

How I use AI (part deux)

Travel-related

Cancun

Day 1

How I use AI

What I’ve used AI for

Laziness

Writing code to use features that I know exist, but am not proficient at:

Working with cumbersome CLI commands

Sex and gender

The art of watching sports