atharva

went to lobste.rs after a while. I think they are still not fans of AI.

analogies are like wet balloons, they are not very effective at explaining things

My colleague analysed the system prompts for codex and claude and realised the reason they feel different is because of deliberate product decisions in the prompt! blog.nilenso.com/blog/2026/02...

things I have dumped on the internet this month. How the lobsters algorithm works: atharvaraykar.com/lobsters/ 11:59 PM: atharvaraykar.com/reinforce/

I wrote a post looking into multiple SWE/coding benchmarks. Many of them measure something narrower than what their names suggests. blog.nilenso.com/blog/2025/09...

Wrote about units of work being a useful lever for getting good results from AI-assisted coding. blog.nilenso.com/blog/2025/09...

I've been poking Srihari, our most experienced engineer @nilenso.com to share his hard-earned knowledge for the benefit of others. Even if you're not an engineering leader like me, this checklist gives a lot of insight into what makes a great engineering org. blog.nilenso.com/blog/2025/09...

grateful for my low-quality education where teachers could not correctly explain the basics of anything right it pushed me to cultivate good epistemics and build truth-seeking habits early on it also forced me to never trust authoritative figures, many are very incompetent

Question for the "launch many parallel claude agents to go faster" crowd: Why isn't that the same problem as the Brooks's law all over again?

despite all the AI hype there's only been two kinds of tools with direct improvements to my life: - chat (like chatgpt, claude ai) - terminal coding agents (like claude code) surprising how nothing better has stuck yet

Several people ask me about how I'm keeping up with all the AI things and finding signal in this noisy landscape. I wrote a guide explaining this. blog.nilenso.com/blog/2025/06...

I was seeing @karpathy.bsky.social's Software 3.0 talk, as one should. Surreal to see him recommend my writing to make a point on AI-assisted coding "Is that me on TV??" moment

I loved it! simonwillison.net/2025/Jun/10/...

Quality, and the cutlets of enthusiasm atharvaraykar.com/quality-and-...

I have revamped my website of essays. it's now more searchable, supports commenting and also doubles as a newsletter. now back to writing again.

is "cracked" the new codeninja/rockstar?

Buddhist Teacher Geshe Langri Tangpa (circa 1100 AD) 🤝 XXXTentacion (2018)

General consensus in the replies and quotes of this seems to be that the entire concept of "AI skills" is a joke - how hard is typing text into a chatbot, really? I will continue to argue that it's genuinely difficult, and that the challenge in using these tools is widely underestimated

the reason i write and encourage other people to write is because writing is the most scalable form of persuasion and negotiation it's one of the best coordination mechanisms that humans have

my current pet conspiracy theory is that a lot of your favourite artists are heavily using AI generated sequences in their music, but no one has noticed it yet

TIL about the Modi (मोडी / 𑘦𑘻𑘚𑘲‎) script, used for Marathi up until the 1940s. In 2025, it looks like AI generated Devanagari.

One of our engineers wrote a thoughtful article about how public health challenges can be framed using "Systems Thinking" @atharvaraykar.com www.simple.org/blog/a-syste...

i picked up india after gandhi by ramachandra guha recently and it is a 10/10 book so far it has renewed my fascination with india context is that which is scarce. this book will catch you all up with some context

i figured out how to use twttr without frying my brain https://buff.ly/3DIRL7W

wrote a short note on why i approach everything with optimism tldr; - optimism is the ultimate form of ownership - cynicism is the rejection of ownership - hard to be curious when you're always cynical

'tapestry' is another linguistic casualty of LLMs

"god is in the details" unexpectedly resonant phrase from conan o'brien and a common thread tying most things i look up to

thindi places get a bad rep sometimes for being "dirty". many aren't, they just don't match your aesthetics of cleanliness i'd still rate those kitchens as multiple levels cleaner than the avg bangalore tech person bachelor pad kitchen

I *occasionally* see complaints here about far-right trolls and toxicity on the twttr, and I complaints on the twttr about far-left virtue signal moralistic trolls on here. I experience neither on either—I'm either weirdly good at curating my feed or not famous enough?

menu highlights

art in a white wall big box chamber is gazed upon for long, an object of loving attention art outside the door of hae kum gang korean restaurant gents toilet is ignored among the din of swinging doors and flushing toilets

chai darkening ideas - add fine mud - spray spray tan - mix in water from a lake in peenya - chocolate - blend with chyawanprash

bluesky has no gifs bluesky has no support for 1834 x 1048 video ew

in case you missed it, the current thing I'm doing with my life is making esoteric thindi hopping videos youtu.be/nZfKcNHs0SI?... #bangalore ?? #dosa ?? #idly ????