Skip to content
View JayCheng113's full-sized avatar
3 followers · 1 following
  • University of Wisconsin Madison
  • Madison
  • 06:05 (UTC -05:00)
  • LinkedIn in/zhan-cheng

Achievements

Achievement: Pair ExtraordinaireAchievement: Pull SharkAchievement: QuickdrawAchievement: Starstruck

Achievements

Achievement: Pair ExtraordinaireAchievement: Pull SharkAchievement: QuickdrawAchievement: Starstruck

Block or report JayCheng113

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 supported. This note will be visible to only you.
Report abuse

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

Report abuse
JayCheng113/README.md

Typing SVG

LinkedIn  Website  Google Scholar  Views

Interested in Reinforcement Learning and AI Agents — building systems that learn, reason, and act.

⚡ Featured Projects

InfiEpisteme

Automated research pipeline generating peer-review-quality papers via Claude Code. 10-stage pipeline with human checkpoints.

stars python

AgentInfra

File-based knowledge system giving AI agents persistent, project-specific understanding via structured .ai/ directory. Zero infra.

stars markdown

Nano-LLaDA

A ~30M param discrete diffusion LM combining AR and diffusion generation. LLaDA 1.0 & 2.0 with iterative remasking sampling.

stars python

EZVibe-Light

Lightweight terminal UI for managing multiple Claude Code sessions. 43KB bundle, zero native deps, keyboard-driven.

ts

Pinned Loading

  1. InfiEpisteme InfiEpisteme Public

    InfiEpisteme 📜 — Talk to Claude Code, get a research paper. 19 Markdown skills orchestrate the full pipeline: literature survey → ideation → experiments → writing → cross-model review. No framework…

    Python

  2. AgentInfra AgentInfra Public

    AgentInfra 🧠 — Give your AI agent a memory that lives with the code. Drop a .ai/ directory of structured Markdown into any repo: architecture → decisions → conventions, three-layer loading at ~3% c…

    1

  3. Nano-LLaDA Nano-LLaDA Public

    Python 86 9

  4. EZVibe-Light EZVibe-Light Public

    TypeScript

  5. skill-retrieval-mcp skill-retrieval-mcp Public

    Python

  6. skill-finder skill-finder Public

    Python