Stepan Goncharov
2.3K posts
Stepan Goncharov
@stepango
Agents & DevProd & Supercomputing @xAI
Bay Area Katılım Ağustos 2009
627 Takip Edilen922 Takipçiler
English
Once you go Tesla you never go back 😎
Wes@wmorrill3
often hear people say other vehicles are ruined after driving a Tesla. The data seems to agree.
English
English
English
> engineering is art
> picture of turbine
Everytime
Saint Nomad@Saint_n0mad
engineering is art
English
1. DataCenter, every SF resident gets a slice of free compute
2. Use it to heat the sutro baths
English
Obviously unbelievable but specifically I'm wondering why they call it ARR. Tokens in enterprise are billed on usage not subscription.
Tannor Manson@Futurenvesting
Anthropic is now showing off $44 BILLION in annual recurring revenue. This is up $14 billion (+46.6%) since last month! BULLISH for AI Infrastructure $NVDA $AMD
English
If your answer is not "same" or "more", I have some bad news for you
@levelsio@levelsio
Do you work less, same or more with AI vs before AI?
English
@thdxr Async image loading/resizing should be a part a platform
English
as usual bunch of people saying "this is bloat just use a library"
image resizing particularly benefits from native code and libraries like sharp are always a pain to deploy into all environments because you need to build the native part
Bun@bunjavascript
In the next version of Bun `Bun.Image` - fast builtin multi-format image processing library
English
Tesla Self-Driving 14.3.2 is rolling out
Jordan Camina@hletes
14.3.2 is ROLLING OUT TO 14.2.2.5 people! Just pulled into a supercharger and I have the update!
English
English
English
@cto_junior Efforts should be spent on learning how to put the right rails for agents, not on reviewing every line of its output
English
If you’re starting to feel like the bottleneck while using Codex or Claude Code,
Congrats!!
You might be in the top 1% of devs who survive the next wave
Xylon.Ai@Xylon_lew
honestly the biggest unlock was stopping trying to hold it all in my head I keep a running doc — just bullet points of what each task needs before I context-switch. sounds basic but it cut my "wait where was I" moments in half the tools are fast enough now that the bottleneck is always my own organization, not the AI
English
what the fuck do you mean "94ms median time-to-show"??
is it making an API request everything i want to open a program?
Windows Latest@WindowsLatest
Microsoft confirms Windows Run dialog is getting a modern look on Windows 11! - Modern design: A refreshed look that matches Fluent Design and Windows 11, with dark mode support. - Faster than before: Perf was top-of-mind when rewriting Run, and with a **94ms median time-to-show time it’s faster than ever before. - Quick user directory access: You can now type ~\ to jump to your user directory, then keep navigating just like you would from the command line. The modern Run dialog is slowly rolling out in current Insider builds of Windows as an opt-in feature, ensuring we are collecting your feedback. To enable the new Run dialog, you’ll need to: - Be on Windows Insider Experimental Channel. - Enable to Settings -> System -> Advanced and toggle on the new experience with the “Run Dialog” option at the top of the screen.
English
@meowbooksj Spot on! Engineers who had management experience are much faster to adapt to new reality, hardcore engineers mostly are not willing to accept the fact that code the agent writes does not look exactly like theirs 😁
English
you think managers read and understand all the code their developers write? no - they delegate that to others for lack of time
you still think 80% of human reviewers are better than 20% of SoTA LLMs?
what you have described is the last bottleneck in software abundance. bye!
English
@rosmine One of the reasons why I like functional programming, it pushes edge cases into side effects which is much easier to manage and makes code easier to reason about, at least for a human software engineers, still need to test how good the agents are with it though
English
Jevon's Enshittification:
The better AI gets, the more code you can write.
Bugs scale linearly with the amount of code, so software quality gets worse instead of better.
Rosmine@rosmine
My day so far: - Security update local gpu server. It does "kernel panic" and fails to boot. 1 hour debugging - Apartment complex has migrated to new payment system. Payment fails. 30 min debugging - Laptop is so old it can't play youtube. Buy a new one and transfer (actually yesterday) - New laptop can't log in. 1 hour debugging so far. Still not solved. I'm spending so much time debugging other peoples bugs that I don't have any time to debug the issues I caused myself
English