Live data from Hacker News

Viewing profile — Raywob

Raywob

HN member
Joined
Sun, Feb 15, 2026, 6:50 AM UTC
HN karma
4
Public activity
12 items

About Raywob

Building zero-cost AI infrastructure on consumer hardware. Dual-model orchestration — local GPU for the grunt work, cloud API for the thinking

Recent public activity

  1. story
    Show HN: Vibe Check – Client-side invisible Unicode steganography scanner

    Glassworm has hit 400+ repos across GitHub, npm, and VS Code using invisible Unicode characters to encode executable payloads that pass every code review, linter, and AI assistant.…

  2. story
    My AI didn't misread a receipt – it fabricated one from scratch

    I pointed a vision model at a grocery receipt. It returned a store name, item list, and total. None of it was on the paper. This wasn't OCR error. The model didn't confuse a "7" fo…

  3. story
  4. story
  5. comment
    Comment #47075227

    Haven't tried GPT-OSS-20B yet — the MOE approach is interesting for keeping VRAM usage down while getting better reasoning. 85 t/s on a 3060 is impressive. I'll look into that. I'v…

  6. comment
    Comment #47074605

    For the mechanical stages (scanning, scoring, dedup) — indistinguishable from proprietary models. These are structured tasks: "score this post 1-10 against these criteria" or "extr…

  7. story
    The $2k Laptop That Replaced My $200/Month AI Subscription

    Cloud AI pricing is per-token. The more useful your pipeline, the more it costs. I built a dual-model orchestration pattern that routes 80% of work to a free local model (Qwen3 8B …

  8. comment
  9. comment
  10. comment
  11. comment
  12. story