1. X
  2. Elliot Chen πŸ¦β€β¬›
Log inSign up
Elliot Chen πŸ¦β€β¬›
606 posts
Elliot Chen πŸ¦β€β¬› profile banner
@elliotchen200

Elliot Chen πŸ¦β€β¬›

@elliotchen200
Building OSS Raven & EverOS at @EverMind. Working with open-source AI communities worldwide. Agents, memory, evals, and deep research.
github.com/EverMind-AI/ra…
Joined February 2011
546
Following
342
Followers
RepliesRepliesRepostsRepostsMediaMediaArticlesArticles

Log in or sign up for X

See what’s happening and join the conversation

Continue with phone
or
Log in with username or email
TermsΒ·PrivacyΒ·CookiesΒ·AccessibilityΒ·Ads InfoΒ·Β© 2026 X Corp.
  • @elliotchen200
    Elliot Chen πŸ¦β€β¬›
    @elliotchen200
    Aug 5
    AI things will get interesting when you see these two top tweets on your timeline. 😏
    Image
  • @elliotchen200
    Elliot Chen πŸ¦β€β¬›
    @elliotchen200
    Aug 4
    Article cover image
    Article
    The Young Builders Shaping AI
    I watched Silicon Valley 101's long interview with Ying Sheng because I wanted to understand the story behind SGLang. Nearly two hours later, I was still thinking about the way she described the work:...
  • @elliotchen200
    Elliot Chen πŸ¦β€β¬›
    @elliotchen200
    Aug 3
    Guess what. Qwen3.8-Max should already work with Raven πŸ¦β€β¬› Raven has native DashScope support plus custom OpenAI-compatible endpoints, including streaming, tool calls, and reasoning_content. github.com/EverMind-AI/ra…
    Image
    Image
    @Alibaba_Qwen
    Qwen
    @Alibaba_Qwen
    Aug 3
    πŸ“’Meet Qwen3.8-Max β€” our most capable model to date. Next week, the open weights of Qwen3.8-Max will be released, and Qwen3.8-27B is also going open-weights to meet you all!πŸŽ‰ Qwen3.8-Max, a new bar for coding and cowork at 2.4T parameters: - Autonomous coding: 10+ days of
  • @elliotchen200
    Elliot Chen πŸ¦β€β¬›
    @elliotchen200
    Jul 31
    The LLM can propose changes. But deterministic evals have to decide what deserves credit. That diagnose-and-credit loop is what turns self-improvement from a demo into a system you can trust.
    @evermind
    EverMind
    @evermind
    Jul 30
    Our new paper. Self-evolving agents have a measurement problem. The hard part isn’t generating changes to an agent harness. It’s knowing which changes genuinely helped rather than got lucky, overfit, or never activated at all. LLMs diagnose failures and propose changes across
    Image
  • @elliotchen200
    Elliot Chen πŸ¦β€β¬›
    @elliotchen200
    Jul 28
    K3’s β€œ3” seems to carry a second meaning: δΈ‰η”ŸδΈ‡η‰©, β€œthree gives rise to all things.” The film moves from the Dao β†’ one β†’ two β†’ three β†’ all creation, then lands on open weights. K3 is the β€œthree.” The community gets to create the rest.
    @Kimi_Moonshot
    Kimi.ai
    @Kimi_Moonshot
    Jul 27
    Kimi K3 (open weights, coming soon)
    Image
    00:00
Advertisement
Advertisement