Vignesh Raja

@DotVignesh

ML @ServiceNow Core AI

dotvignesh.github.io

Joined June 2018

Replies Reposts Media

Pinned
Vignesh Raja
@DotVignesh
Aug 31
On ITBench, Qwen3.5-27B with an evolved harness beats GPT-5.5 on the baseline harness by +19.2 points. More capable models do not always make better enterprise agents. With the right harness, a smaller open-weight model can outperform a larger frontier model running the default
Vignesh Raja
@DotVignesh
Jul 27
“The idea that you’ll be in the permanent underclass if you don’t join a frontier lab is so dumb” -@sama at YC SUS 😂
00:00
Vignesh Raja
@DotVignesh
Jul 26
“Apparently, if you’re here, you’re going to make it. So I’m happy I’m here.” - Jensen at @ycombinator AI Startup School
00:00
Vignesh Raja
@DotVignesh
Jul 24
Despite having a model as good as Fable, Anthropic still uses haiku 4.5 for creating their charts…
Claude
@claudeai
Jul 24
Replying to @claudeai
On several coding and knowledge work evaluations, Opus 5 is the new state-of-the-art:
Vignesh Raja
@DotVignesh
Apr 3
Not just emotions, but LLMs also have concept vectors representing their propensity to take certain actions. In our recent work, we tried to answer the question: when a reasoning model makes a choice, did it think first and then decide to, or decide first and then think? 🧵
Denis A. and Sai Rajeswar
Anthropic
@AnthropicAI
Apr 2
04:52
New Anthropic research: Emotion concepts and their function in a large language model. All LLMs sometimes act like they have emotions. But why? We found internal representations of emotion concepts that can drive Claude’s behavior, sometimes in surprising ways.
2