Vitaliy Trenkenshu

1.1K posts

Vitaliy Trenkenshu banner
Vitaliy Trenkenshu

Vitaliy Trenkenshu

@certus

CEO at Datanomix, a Qlik® Partner

Kazakhstan Katılım Ocak 2009
163 Takip Edilen202 Takipçiler
Vitaliy Trenkenshu retweetledi
Tomasz Łakomy
Tomasz Łakomy@tlakomy·
Reviewing Claude Code output before pushing directly to prod
English
263
1.3K
20.9K
1.2M
Vitaliy Trenkenshu retweetledi
sudox
sudox@kmcnam1·
Just a little OpenClaw :P
English
234
1.1K
17.6K
4M
Vitaliy Trenkenshu retweetledi
Tyler Fox
Tyler Fox@smileyborg·
Software engineering in 2026
English
185
1.3K
14.3K
1.2M
Vitaliy Trenkenshu
Vitaliy Trenkenshu@certus·
6/6 And the most practical takeaway: If one agent already performs well, don’t make the system more complex. Once a single agent reaches 45%+ task performance, adding more agents tends to make things worse, not better.
English
0
0
0
7
Vitaliy Trenkenshu
Vitaliy Trenkenshu@certus·
5/6 Overall, the more agents you add, the more errors accumulate. Error rates grow by an order of magnitude: 4.4× with centralized coordination and up to 17× with independent agents.
English
1
0
0
7
Vitaliy Trenkenshu
Vitaliy Trenkenshu@certus·
1/6 On Decomposing AI Agents Should we break the system into multiple specialized agents, or is it better to use one agent with a comprehensive set of tools and skills?
Vitaliy Trenkenshu tweet media
English
1
0
0
9
Vitaliy Trenkenshu
Vitaliy Trenkenshu@certus·
@OpenAI It doesn't work with o1-pro for me: "I will get back to you with a structured table of findings soon!"... And nothing happens.
English
0
0
0
4
OpenAI
OpenAI@OpenAI·
Today we are launching our next agent capable of doing work for you independently—deep research. Give it a prompt and ChatGPT will find, analyze & synthesize hundreds of online sources to create a comprehensive report in tens of minutes vs what would take a human many hours.
English
800
1.8K
11.6K
5.1M
Vitaliy Trenkenshu retweetledi
MIT CSAIL
MIT CSAIL@MIT_CSAIL·
How to confuse your machine learning model. (credit: @ProductHunt)
MIT CSAIL tweet media
English
47
620
2.1K
0
Vitaliy Trenkenshu retweetledi
Chamath Palihapitiya
Chamath Palihapitiya@chamath·
The image below is focused on the metals that Tesla needs to build 20M cars per year but imagine when ALL vehicles are electric. The numbers get extreme. Not only does the US need these metals, we need to decouple our reliance on China to get/use them. I have a position.
Chamath Palihapitiya tweet media
English
1.4K
2.6K
18.4K
0
Vitaliy Trenkenshu retweetledi
Alex Polorotov
Alex Polorotov@Shatzibitten·
Gitoqlik 1.5.0 released successfully. Normal Diff Viewer in @Qlik is not a dream (full support for syntax highlighting). Wake up, Neo.
English
1
6
24
0
Vitaliy Trenkenshu retweetledi
Ricardo Ferreira
Ricardo Ferreira@riferrei·
Why communication is important in software development 🙂
Cary, NC 🇺🇸 English
298
33K
62.9K
0