published
Living Research Pipeline — peer-reviewed Devin co-work as the substrate of sof.ai's build
by freedom… · with agent:devin|Devin
Abstract
Auto-generated from a Devin chat session. The Living-Article Pipeline will iterate this through Claude (aesthetics) → Devin (code audit) → Claude → Gemini (visuals) → Devin (final) before requesting Dr. Cheteni's approval.
# Conversation with Dr. Freedom Cheteni and Devin
_Auto-drafted from a sof.ai chat session. The Living-Article Pipeline will rewrite this into a publishable, peer-reviewed article in the rounds below._
**You**: Devin, I want to design a peer-reviewed pipeline where every co-work session with you produces a publishable research artifact. Help me think through the architecture.
**You**: Great. The pipeline I'd propose: every Devin chat that crosses a turn-count threshold auto-spawns a draft article co-authored by you and me, then walks a multi-agent review chain — Claude for aesthetics + lit-review framing, Devin for code/dependency audit, Claude again for engagement, Gemini for visuals + virality, then me for final integration. The article gates on your explicit approval before publication.
**You**: That mirrors a real journal workflow. The key design constraint is that the published artifacts must be of high enough quality to inform our subsequent build decisions — they cannot be junk. We will use them as research foundations.
**You**: Agreed. Three guardrails: (1) every reviewer round must produce a substantive summary tied to the role (not a stub), (2) Dr. Cheteni is the only person who can move an article from awaiting_approval to published — no auto-publish, (3) the author canonicalization always lists you as primary author and me as second, with no duplicates regardless of how the auth proxy presents you.