Professor of Medical AI @ SYSU
Building clinically grounded AI, one-person-lab workbenches, domain agents, and auditable workflow infrastructure.
|
Academic Site fenggaolab.org |
OPL ShellOne Person Lab
|
Medical Domain AgentsMed Auto ScienceMed Auto Grant
|
Visual Domain AgentRedCube AI
|
This profile serves as a public entry point for my academic and technical work in medical AI, research systems, and clinically grounded workflow design.
The current public surfaces are:
fenggaolab.orgfor academic informationOne Person Labas the top-level GUI and management shell for a one-person research labMed Auto Scienceas the active medicalResearch Opsdomain module underResearch FoundryMed Auto Grantas the active medicalGrant Opsdomain module underGrant FoundryRedCube AIas the active visual-deliverable domain module underPresentation Foundry
One Person Lab is the top-level GUI shell, management surface, family gateway, and module catalog for a one-person research lab.
It currently owns:
- the public entry shell for everyday collaboration, general tasks, and specialized workflows
- the module catalog for active domain agents and future workstreams
- workspace-level progress feedback, file delivery, settings, and module management
- family-level naming, shared helper language, and machine-readable handoff contracts
- top-level routing and federation language across current and future domain modules
Domain modules keep their own canonical domain truth, audit writeback, runtime surfaces, and delivery records.
Human
-> One Person Lab shell
-> Everyday Codex conversation
-> General Codex tasks
-> Research Foundry -> Med Auto Science
-> Grant Foundry -> Med Auto Grant
-> Presentation Foundry -> RedCube AI
-> Progress / Files / Settings / Module Catalog
Default execution:
Domain module -> Codex default execution
Online supervision:
Explicit long-running lane -> Hermes-Agent gateway
| Product family | Active module | Public role | Main outputs |
|---|---|---|---|
Research Foundry |
Med Auto Science |
Medical Research Ops domain module |
Study workspaces, evidence packages, manuscripts, submission materials |
Grant Foundry |
Med Auto Grant |
Medical Grant Ops domain module |
Grant directions, proposal drafts, review packs, submission-ready packages |
Presentation Foundry |
RedCube AI |
Visual-deliverable domain module | Slide decks, scripts, reviewable visual artifacts, export bundles |
Future workstreams such as Thesis Foundry and Review Foundry stay in definition until their domain package and public boundary are ready.
Med Auto Science is the current active medical implementation on the Research Foundry line under the One Person Lab umbrella.
It focuses on medical research operations, including:
- disease-workspace organization
- data governance
- study progression
- evidence packaging
- manuscript and submission delivery
Med Auto Grant is the active medical grant-writing implementation on the Grant Foundry line.
It focuses on author-side grant operations, including:
- direction screening
- question refinement
- argument building
- proposal drafting
- critique, revision, and package export
RedCube AI is the active visual-deliverable implementation on the Presentation Foundry line.
It currently focuses on:
ppt_deckand presentation-grade visual delivery- visual deliverable governance and review loops
- agent-first production of audited visual artifacts
Together, the current public shape is:
One Person Lab -> Research Foundry -> Med Auto Science
alongside:
One Person Lab -> Grant Foundry -> Med Auto Grant
alongside:
One Person Lab -> Presentation Foundry -> RedCube AI
My current work is centered on:
- clinically grounded AI for colorectal cancer
- agent-first, human-auditable domain systems
- medical research workflow design
- one-person-lab infrastructure for research, grant, and visual delivery work
- Academic site: fenggaolab.org
- One Person Lab: gaofeng21cn/one-person-lab
- Med Auto Science: gaofeng21cn/med-autoscience
- Med Auto Grant: gaofeng21cn/med-autogrant
- RedCube AI: gaofeng21cn/redcube-ai