Shameer C

Architect, Distributed Systems || Scala, PHP, Javascript || Linux, AWS || Director of Engineering @careem || Writer @sitepointdotcom & @Dzone || @phptodayorg

[Omar Zhan @OMAR_JAHN]

@lennysan when the cost of building drops to zero, the bottleneck moves to knowing what's worth building. ambition matters because that's the part that doesn't get cheaper. anyone can ship now. almost nobody can pick the right hard thing and stay on it long enough for it to matter.

[Mitchell Hashimoto @mitchellh]

I've got an agent in a loop optimizing a renderer with the goal to minimize frame times (and tests to measure). It got times down from 88ms to 2ms and allocations down from ~150K to 500. Sounds good, right? Wrong. This is exactly why agent psychosis is a big fucking problem.

As an experiment, I rewrote the Ghostty core render state in Go, with access to identically laid out data structures as Ghostty and the exact same validation tests. I made a purposely naive renderer (simple, correct, but slow). 88ms per frame with 150,000 allocations (horrendous, lol)!

I then kickstarted a Ralph loop to bring the frame times down. I told it it can't modify input data structures or the public API or tests (they're correct), but it can do anything else it wants. It got to work.

It has worked for about 4 hours. I've spent around $350 on this experiment so far. The results?

88ms => 1.5ms
150K allocs => ~500 allocs

Incredible right? Nope.

My hand-written renderer I ported has frame times (same benchmark) of ~20us (0.020ms) and 0 allocations in the update path.

This is the problem with psychosis and lacking systems understanding. If you don't understand the system, you're going to accept that this is an incredible result. If you understand the system, you'll see better solutions immediately and can do roughly 75x better on throughput.

The people who blindly trust agent output are in the former camp. They're sheeple, overdrinking from a fountain of mediocrity.

Standard disclaimer: I use AI all the time. I like AI. The point I'm making is to not blindly accept results. Think. Analyze. Learn.

[Shameer C @shameerc]

Honestly I didn’t know it’s what in everyone else’s mind.

[Akash @kaaaash____]

Be honest,

Why is every software engineer’s backup plan always farming?

[Cormac @cormachayden_]

software engineers before vs after agents

[Boring_Business @BoringBiz_]

CFOs realizing that their AI token budget is going to be higher than the salaries of the people they laid off

[Shameer C @shameerc]

@AmazonHelp Please share a new link

[Shameer C @shameerc]

. @amazonIN Your service is absolute garbage with orders getting cancelled or indefinitely delayed without any visibility of the reason, updated eta and more importantly contact someone to get an update. Your delivery team marks attempted delivery without even contacting. @amazon

[Shameer C @shameerc]

@AmazonHelp @AmazonHelp Sorry, I am done with ordering and not taking any further effort. Contact me if you want my feedback and improve your services.

[Shameer C @shameerc]

What an epic ad.. 👏👏

[Armin Ronacher ⇌ @mitsuhiko]

Everything is crazy. The real super power is somehow not getting affected by it.

[Andrej Karpathy @karpathy]

The hottest new programming language is English

[Shark @fintech_shark]

This guy predicted a year ago that Trump would win 2024 election and attack Iran, instigated by Netanyahu.

Bro teaches the history even before the actual events happen. Very very insightful

[Ryan Els @RyanEls4]

It works on my machine 😂

[Shameer C @shameerc]

@GergelyOrosz Interviewer takes questions from llm
Candidate answers with the help of llm
Interviewer asks llm to score the candidate
Interview death cycle

[Paul Couvert @itsPaulAi]

Hugging Face released a free course on agents

You can learn how to create:

- Code agents that solve problem with code
- Retrieval agents that supply grounded context
- Custom functional agents that do whatever you need

Definitely the best I've seen so far.

Link below

[Shameer C @shameerc]

Spot on!

[Shreya Shankar @sh_reya]

another problem with LLMs as SWEs is that they don’t know when to refactor. a good example of this is when using react contexts vs hooks. When you instruct LLMs what to do, they don’t have the forward-looking vision of the app that you have, and it’s almost always more optimal to simply add one more piece of state to the context than refactor into hook(s)

[Barry Malone @malonebarry]

I feel utterly sick. There are no words.

[sanjoy ghose @advsanjoy]

That anti-national Wire has made this video! Chi chi. Very bad 😬

[Crémieux @cremieuxrecueil]

This is not good.

Hachette Books, Penguin Random House, HarperCollins, and Wiley have won their case and their books are going to be removed from the Internet Archive.

If you still want or need to get their books for free, use libgen and Anna's Archive.

[Waseem ವಸೀಮ್ وسیم @WazBLR]

This bengaluru auto rickshaw driver knows what's he talking about.