Sergei Vecherenko

635 posts

Sergei Vecherenko banner
Sergei Vecherenko

Sergei Vecherenko

@serg_vecher

Building https://t.co/txo1elNnVc | Architecting AI Search 🤖 | Lead Frontend Engineer | Sharing insights on engineering leadership & AI workflows | CA 📍

California, US Katılım Mayıs 2011
91 Takip Edilen84 Takipçiler
Sabitlenmiş Tweet
Sergei Vecherenko
Sergei Vecherenko@serg_vecher·
Lightcode 1.1.0 is out. Now with: • Grok Build, install from app and run • Claude Opus 4.8 with Ultracode effort and workflows • Local voice input • Agents One-click install • Thread rollback to undo an agent's work and retry • Live GitHub PRs in-app • Hook plugin install and uninstall for CLI Agents • And more then 70 PRs with more and fixes Bring your own subscriptions and keys. macOS, Windows, Linux. lightcodeapp.com github.com/SDSLeon/lightc…
Sergei Vecherenko tweet media
English
0
0
1
43
Sergei Vecherenko
Sergei Vecherenko@serg_vecher·
@OpenAI Did you fix Sandbox on Windows so it will not corrupt ASL? I am scary to install Codex on Windows...
English
0
0
0
682
OpenAI
OpenAI@OpenAI·
Windows users, this one’s for you. Computer use now works on Windows, so Codex can take action on your Windows computer. And with Windows support for Codex in the ChatGPT mobile app, you can start, review, and steer tasks on the go while work continues on your Windows machine. An early experience, but we’re working on more ways to keep your work moving, wherever you are.
English
783
884
8.3K
1.2M
Sergei Vecherenko
Sergei Vecherenko@serg_vecher·
@rezoundous I have both (200$ Claude and 100$ Codex) it will not give you smt better then gpt-5.5, I would say now it's just better then before, maybe on pair with gpt-5.5 high. If you already have 200$ Codex and want something to more, then you should subscribe.
English
0
0
0
44
Tyler
Tyler@rezoundous·
Should I sub claude code again? Fomo is real.
English
35
0
55
8.9K
Sergei Vecherenko
Sergei Vecherenko@serg_vecher·
Everyone so exited for opus 4.8 today, like they will not cry about limits tomorrow, and that it's degraded in 1 month.
GIF
English
0
0
0
37
Tyler
Tyler@rezoundous·
Seems like I'm on track to work 4 days a week, and touch grass for the other 3.
Tyler tweet media
English
14
0
56
3.9K
Sergei Vecherenko
Sergei Vecherenko@serg_vecher·
Codex compaction is best on the market, you just don't care about context, it's just works. With 258k. In other side Opus 1m where compaction is like another task where you have time to get some drinks. Big context start working bad, but you don't want to compact, loop.
English
0
0
0
19
Luke Parker
Luke Parker@LukeParkerDev·
its not vertically centered 🫠
Luke Parker tweet media
English
1
0
5
1.1K
Sergei Vecherenko
Sergei Vecherenko@serg_vecher·
This week Codex limits burn way faster, can't imagine, what will be after promotion ends...
Sergei Vecherenko tweet media
English
0
0
0
38
Daniel Smidstrup
Daniel Smidstrup@DanielSmidstrup·
Which command do you use most in the terminal?
English
361
4
142
43.4K
Sergei Vecherenko
Sergei Vecherenko@serg_vecher·
lightcodeapp.com will notify you when new agents version are available and you can update them with 1 click. While typing your prompt! Claude Code, Codex, Gemini, Grok Build, Copilot, Open Code no matter what agent are you using or all at once! Get latest version here: lightcodeapp.com/nightly or stable here: lightcodeapp.com
Sergei Vecherenko tweet media
English
0
0
1
82
Daniel San
Daniel San@dani_avila7·
/simplify is back in the latest Claude Code update It's now an alias for /code-review --fix Which applies review findings to your working tree after the review, surfacing reuse, simplification, and efficiency suggestions Going to be testing it over the next few days, will report back on how it goes 🫡
English
32
23
387
46.2K
Sergei Vecherenko
Sergei Vecherenko@serg_vecher·
Gemini Models state of tool calls 2 times tried to edit not existed index.html then created it with cat / node 😒
Sergei Vecherenko tweet mediaSergei Vecherenko tweet media
English
0
0
0
15
Sergei Vecherenko
Sergei Vecherenko@serg_vecher·
@serenaa_ge Why different efforts and not most used ones? gemini-3.5-flash - medium (should be high) gpt-5.5 - xhigh (shoud be high) claude-opus-4.7 - max (should be xHigh)
English
1
0
2
1K
Serena Ge (Datacurve)
Serena Ge (Datacurve)@serenaa_ge·
Today we’re releasing DeepSWE, a new standard for agentic coding benchmarks. On public leaderboards, top models often look relatively close in capability. DeepSWE shows where they actually diverge, reflecting the realistic experience of developers in their day-to-day work.
Serena Ge (Datacurve) tweet media
English
502
734
5.9K
1.9M

Keşfet

@OpenAI @rezoundous @LukeParkerDev @DanielSmidstrup @dani_avila7 @elonmusk @BarackObama @taylorswift13