Vijay Vankayalapati

Data Engineer | Golang & Rust Enthusiast | Formula 1

[Guillermo Rauch @rauchg]

If you’re not reading the code, whether explicitly or through agentic inquiry, one or more of these is true:

○ You’re a beginner
○ Software is throwaway
○ You’re prototyping
○ You have no users / revenue
○ You’re taking on debt & risk
○ Your problems are basic

And btw. All of this is fine. But the reality is that models are still not at the “full autonomy” stage yet.

They make rookie mistakes, they go down bad architectural paths. I just had the best model in the world add a nonsensical 700ms delay to “settle” something and it told me “you’re right, I was cargo-culting” 🤨

I am on the camp that this need will diminish more and more. Most code is indeed going to be assembly-like. But we also have the global internet and software infrastructure riding on these models and narrative, and we have to respect that.

[Mitchell Hashimoto @mitchellh]

Btw this is why computer science matters. Yes, you can self teach it if you really want to (I personally wouldn't have had the discipline, so college was a way better force). However you learn it, the theoretical fundamentals matter and get applied to work every single day.

[slash1s @slash1sol]

A GOOGLE ENGINEER STOOD UP AT A C++ CONFERENCE AND TOLD THE ROOM THAT THE DATA STRUCTURE YOU REACH FOR BY REFLEX -- THE LINKED LIST -- IS ALMOST ALWAYS THE WRONG ANSWER. NOT SLOWER BY A LITTLE. SLOWER BY SO MUCH THAT USING IT IS QUIETLY THROWING AWAY YOUR HARDWARE

73 minutes from Chandler Carruth -- the engineer who leads the Clang compiler team at Google.

-> His split: efficiency is how much work you do. Performance is how fast the machine does it. Everyone obsesses over the first and ignores the second.

The textbook says a linked list has "O(1) insert". True on paper. On real silicon it scatters your data across memory and the CPU stalls chasing pointers.

An array looks dumber. Copy the whole thing to insert? Yet it sits in cache, moves in a straight line, and destroys the "smarter" structure in practice.

His point that stings: your Big-O intuition was taught on a machine that no longer exists. The cache, not the algorithm, decides who wins now.

And it's the AI era's silent tax too. Oceans of rented compute stalling on cache misses because the "clean" structure never touches memory the way the chip wants.

You thought choosing a data structure was about theory. This is the man who builds the compiler saying it's about how memory actually moves.

Save this. You'll question every linked list you ever reach for ↓

[Polymarket @Polymarket]

JUST IN: Ford rehires more than 300 veteran human engineers after it says AI failed to deliver the same level of expertise.

[SemiAnalysis @SemiAnalysis_]

Transformer’s Attention mechanism has come a long way. We’d like to thank the researchers and the engineers in the open-source community for continuing to make high-performance AI accessible. Please celebrate with us by sharing this post, tagging more contributors, and sharing anecdotes to complete the open history of Attention! (1/8)🧵

[Strace @straceX]

we named it random access memory (RAM).

then we built three levels of cache, prefetchers, data-oriented design, and an entire performance-engineering discipline whose whole purpose is making sure nobody accesses it randomly.

[Jeff Dean @JeffDean]

People replace their phones every ~4 yrs. This means there are hundreds of millions of old phones discarded each year that are still perfectly usable as computing devices. @Google in collabration with @UCSD is exploring how to turn these old phones into cloud-computing “phone clusters”. Putting phones back in service in this way can directly reduce the environmental footprint of computing by avoiding the need for further raw material extraction, and taking advantage of the embodied carbon already incurred from manufacturing these devices, and modern phones actually are already quite powerful computers. Read more in the blog below ⬇️

[Vijay Vankayalapati @vijayv500]

Opus-4.6 is such a delight to work with.

[Vijay Vankayalapati @vijayv500]

@maninekkalapudi Gotcha 👍

[Vijay Vankayalapati @vijayv500]

@maninekkalapudi perhaps the antidote to this is plans starting at $200/m resulting in permanent underclass with great engineering skills. 😃

[Vijay Vankayalapati @vijayv500]

@maninekkalapudi What are your specs? I have Linux Mint on old lenovo with 4GB ram. Keep posting any interesting things you do with your old machine.

[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.

[Vijay Vankayalapati @vijayv500]

Codex is almost neck to neck with ClaudeCode in plan mode.

[Vijay Vankayalapati @vijayv500]

Artisanal programmers (vs semi-automated ones)

[Joran Dirk Greef @jorandirkgreef]

Something we value at TigerBeetle is that we are craftspeople. It’s who we are and what we do, but more, it’s the quality (and trust) we sell:

“Each line of TigerBeetle code is handcrafted, and then independently reviewed (and understood) by another engineer.”

[Vijay Vankayalapati @vijayv500]

@maninekkalapudi long back saw his name in the citations for some ML paper. Upon looking up he turned out to be a Turing award winner and also first director of iiit-hyd. :-)

[Vijay Vankayalapati @vijayv500]

@maninekkalapudi Dr. Raj Reddy.

[Joran Dirk Greef @jorandirkgreef]

Software is pure “thought stuff”. One person can write code and billions can run it. If anything, our linear time produces exponential value.

Therefore, I’ve never personally believed that developer time is expensive, that we have a “typing” problem. Or that English is somehow a better way to express code than a language as explicit as Zig.

Granted, there’s tons of (non valuable) bespoke software that LLMs can now create. But the valuable thought stuff? Great systems coders are becoming more valuable than ever.

In the land of the blind, the one eyed man is king.

[Vijay Vankayalapati @vijayv500]

Positive step 👍

[ANI @ANI]

#WATCH | The Hague, Netherlands | Tata Electronics and ASML sign an MoU in the presence of Prime Minister Narendra Modi.

(Source: ANI/DD)