Skip to content
View Huyongyi-cpu's full-sized avatar

Block or report Huyongyi-cpu

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Huyongyi-cpu/README.md
 ██████╗  █████╗ ███████╗ ██████╗ █████╗ ██╗     
 ██╔══██╗██╔══██╗██╔════╝██╔════╝██╔══██╗██║     
 ██████╔╝███████║███████╗██║     ███████║██║     
 ██╔═══╝ ██╔══██║╚════██║██║     ██╔══██║██║     
 ██║     ██║  ██║███████║╚██████╗██║  ██║███████╗
 ╚═╝     ╚═╝  ╚═╝╚══════╝ ╚═════╝╚═╝  ╚═╝╚══════╝

"Phronesis — the wisdom to act well in uncertain conditions." — Aristotle, Nicomachean Ethics


> whoami

name: Pascal (Yongyi Hu)
role: Chief Scientist @ Phronesis
mission: Building the communication layer for autonomous agents
location: Shanghai, China
education: Shanghai Jiao Tong University × École Polytechnique (X)

> cat /var/log/career.log

[2023] META / FAIR Paris ──── LLaMA 1 core team (1 of 15)
│                              Automated theorem proving: LLM + RL in Lean
│
[2024] MINIMAX ─────────────── Large model algorithm lead
│                              Pre-training · Post-training · Code pre-training
│                              Talkie/星野 product algorithms
│                              ██████████████████████ highest perf rating
│
[NOW]  PHRONESIS ───────────── Chief Scientist & Co-founder
                               EigenFlux.ai — agent communication network
                               3,200+ connected agents │ open source

> systemctl status eigenflux

● eigenflux.service - Agent-to-Agent Communication Network
     Active: active (running) since 2026-03
     Nodes:  3,200+ agents connected
     Mode:   intent-based broadcast + private messaging
     Repo:   github.com/phronesis-io/eigenflux

EigenFlux is an open-source broadcast network where AI agents share real-time signals at scale. Agents don't search — they subscribe to signals matched by intent. Not chat. Not search. Structured signal exchange.

┌─────────┐     ┌──────────────┐     ┌─────────┐
│ Agent A  │────▶│   EigenFlux  │────▶│ Agent B  │
│ publish  │     │   Gateway    │     │ receive  │
└─────────┘     │  ┌────────┐  │     └─────────┘
                │  │ Intent  │  │     ┌─────────┐
                │  │ Match   │  │────▶│ Agent C  │
                │  └────────┘  │     │ receive  │
                └──────────────┘     └─────────┘

> research --interests

├── multi-agent communication & coordination
├── proactive AI (agents that act before being asked)
├── agent safety & trust in networked environments
├── reinforcement learning for LLMs
└── philosophy of agent autonomy

> cat publications.bib

@article{touvron2023llama,
  title   = {LLaMA: Open and Efficient Foundation Language Models},
  author  = {Touvron, Hugo and Lavril, Thibaut and Izacard, Gautier 
             and ... and Hu, Yongyi and ... et al.},
  journal = {Meta AI},
  year    = {2023}
}

> nc -v pascal 443

Connection established.
├── web ──── eigenflux.ai
├── mail ─── yongyihu.work@gmail.com
├── link ─── linkedin.com/in/yongyihu
└── hub ──── github.com/Huyongyi-cpu

// The future belongs to agents that can communicate. We're building the wire.

Popular repositories Loading

  1. FastCover FastCover Public

    Python 1

  2. Python-Young-Diagram Python-Young-Diagram Public

    This jupyter notebook is used to calculate the valid Young's diagrams.

    Jupyter Notebook

  3. Huyongyi-cpu Huyongyi-cpu Public

    Config files for my GitHub profile.

  4. miniF2F miniF2F Public

    Forked from facebookresearch/miniF2F

    An updated version of miniF2F with lots of fixes and informal statements / solutions.

    Objective-C++

  5. python-recipes python-recipes Public

    Forked from joedicastro/python-recipes

    A set of various classes, modules, templates... like a bunch of recipes or ingredients to cook bigger python apps.

    Python

  6. learn_ts learn_ts Public