Live data from Hacker News

Viewing profile — arizen

arizen

HN member
Joined
Mon, Nov 06, 2023, 12:35 PM UTC
HN karma
244
Public activity
106 items

About arizen

I built deepseek-mcp (github.com/arizen-dev/deepseek-mcp) and run ActiveWizards (activewizards.com) - an AI engineering studio shipping production systems since 2014.

Essays on agent architecture, evaluation, and intelligent organizations: arizenai.com

Production-Ready AI Agents Book (2025): amazon.com/dp/B0G43PPDSP

Contacts: github.com/arizen-dev bobriakov.igor [at] gmail

Recent public activity

  1. comment
    Comment #49121055

    Why not also Qwen?

  2. comment
    Comment #49041391

    Could these complex/hard to read Fable outputs be sign of some kind of industrial level of intelligence, which us humans may have a hard to comprehend, while it may be also hard fo…

  3. comment
    Comment #49002832

    Have you tested Qwen3.8-Max yet?

  4. comment
    Comment #48984294

    Probably increasing learning rate have some value for these players which is enabled by subsidized usage. And cost of acquiring learning via users feedback versus additional learni…

  5. story
    Ask HN: Guidelines on GPT 5.6 Sol steering

    I converged on these principles in my work with Sol so far: Let Sol medium/high propose and implement minimum viable solution without overengineering and overthinking. Use XHigh/Ma…

  6. comment
    Comment #48939498

    And DeepSeek V4 Flash + GLM 5.2 is a really good blend of both (fast/cheap DS + more intelligent GLM)

  7. comment
    Comment #48857232

    Yes, true creativity usually or mostly comes from real constraints, in my experience. As, if there are no constraints in some specific area, there is no kinda "survival need" to im…

  8. comment
    Comment #48849220

    "GPT‑5.6 delivers a step change in design judgment. With only high-level direction, GPT‑5.6 creates tasteful, ergonomic, and functional interfaces. Its stronger computer-use capabi…

  9. comment
    Comment #48848737

    I'm still confused is it available to public via some sort of subscription?

  10. comment
    Comment #48759530

    It was not idiotic from capitalism short-term profit gains point of view.

  11. comment
    Comment #48689803

    Sol Goodman

  12. comment
    Comment #48689774

    Ditto on GLM 5.2 + DeepSeek V4 Flash combo. For most important work (complex, cross-domain inquiries etc.), I still rely on Codex GPT 5.5 though.

  13. comment
    Comment #48539711

    Some anecdata on Fusion: I run same query I used for Fable on OR Fusion and results were worse. It felt, like Fable was able to kinda grasp very deep knowledge/intelligence layers …

  14. comment
    Comment #48539653

    OpenCode Go is pretty good in my experience too. I ended up using DeepSeek V4 Flash as main workload model, while keeping DeepSeek V4 Pro and Qwen 3.7 Plus as advisors on system ar…

  15. comment
    Comment #48502833

    opensourcing software may enable leverage of wider network of contributors to given piece of software,hence software can evolve much more quickly and efficiently.

  16. story
  17. comment
    Comment #48467601

    Probably there will always be frontier surface which frontier model of a given generation would not be able to automate.

  18. story
  19. story
  20. story
  21. story
  22. story
  23. story
  24. story
  25. comment
    Comment #47791322

    Curious how it compares to Browser Use