ChrisWho's played with https://paperclip.ing/ ? Not sure how I wandered across it and I'll get to playing with it eventually, but wondered if anyone had any hot takes on it?
Andrew Lipski
sayzardMicrosoft Research (@MSFTResearch)
AsgardBench는 시각 관찰을 바탕으로 임베디드 에이전트가 작업 중 계획을 수정할 수 있는지 평가하는 벤치마크다. 지각 기반 계획 능력에 초점을 맞춰 에이전트의 한계를 드러내고, 신뢰성 향상에 필요한 개선 방향을 제시한다.
Microsoft Research (@MSFTResearch) on X
AsgardBench evaluates whether embodied agents can revise their plans based on visual observations as tasks unfold. By focusing on perception-driven planning, it exposes key limitations and guides improvements in agent reliability. https://t.co/6jAXzgCLvH
Z.ai for Startups (@ZaiforStartups)
CodeBuddy와 GLM이 싱가포르에서 글로벌 AI 해커톤을 개최한다. 참가자들은 AI 에이전트를 실제로 만들고 배포하며, 무엇이 문제를 일으키는지 확인하는 방식으로 개발을 진행한다. 1,000달러 이상 상금과 멘토링이 제공되며, 4월 20일까지 지원 가능하다.
BOOTOSHI (@KingBootoshi)
한 에이전트가 문제를 해결하지 못하면 다른 에이전트를 추가로 투입해 서로 다른 LLM 모델의 제안을 비교·보완하는 방식으로 더 많은 컴퓨팅을 활용할 수 있다고 소개한다. 여러 모델을 조합해 다양한 해법을 얻는 실용적인 에이전트 활용 팁이다.
BOOTOSHI 👑 (@KingBootoshi) on X
you guys know you can throw more compute at a problem yourself right? if one agent couldn't solve it, throw a diff agent at it who can then see the proposed solutions and offer a variety of different ones BIG help especially when they're different LLM models works everytime!
Jeff BaileyWill agent clusters running agent clusters eat the whole?
#GenAI #AI #Agents #Software #Technology #Programming #SoftwareDevelopment #Coding #SoftwareEngineering
Hacker NewsClaude skill that evaluates B2B vendors by talking to their AI agents
https://github.com/salespeak-ai/buyer-eval-skill
#HackerNews #Claude #B2B #vendors #AI #agents #evaluation #sales #technology
GitHub - salespeak-ai/buyer-eval-skill: B2B software vendor evaluation skill for Claude Code — domain-expert questions, vendor AI agent conversations, evidence-based scoring
B2B software vendor evaluation skill for Claude Code — domain-expert questions, vendor AI agent conversations, evidence-based scoring - salespeak-ai/buyer-eval-skill
sidj79Andrej Karpathy's recent podcast interview is worth your time
Key ideas: agent orchestration over single-session prompting,
AutoResearch loops that remove human researcher from hyperparameter tuning, and a prediction that digital transformation leads while physical robotics lags by years
His take on open source (~6-8 months behind frontier) being a healthy power balance is worth sitting with
"Centralization has a very poor track record." Hard to argue
