Skip to content
View study8677's full-sized avatar

Block or report study8677

Report abuse

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

Report abuse
study8677/README.md

Hi, I'm Jingwen Fan (范敬文) 👋

AI/ML learner focusing on Large Language Models (LLMs).
Passionate about the full LLM lifecycle from pre-training & fine-tuning to alignment (RLHF) and inference optimization.


🚀 Current Focus

  • LLM Alignment: RLHF, PPO, DPO & Retrieval-Augmented Generation (RAG)
  • Learning: Agentic Workflows, DeepSpeed & model quantization (AWQ / GPTQ)

🎓 Background

  • Education: Qilu University of Technology (QLUT)
  • Research Interests: Reward modeling, context window extension, chain-of-thought (CoT)

🎯 Goals

  • Life goal: Stay curious, be brave, and live with kindness.

📟 最新文章

💌 联系方式

Anurag's GitHub stats

Pinned Loading

  1. antigravity-workspace-template antigravity-workspace-template Public

    🪐 The ultimate starter kit for Google Antigravity IDE. Optimized for Gemini 3 Agentic Workflows, "Deep Think" mode, and auto-configuring .cursorrules.

    Python 941 189

  2. easy_claude_code easy_claude_code Public

    Building on prior minimal implementations, this project explains the working principles of Claude Code with fewer core concepts.在前人极简实现的基础上,用更少的概念解释清楚 Claude Code 的工作原理。

    Python 4 1

  3. PromptLint PromptLint Public

    PromptLint — Lint prompts for robustness across models and temperatures.

    Python 5 2

  4. clawdbot-webchat-lite clawdbot-webchat-lite Public

    一个*轻量、可国内部署、多端同步*的 Clawdbot 聊天客户端。不做新 channel,仅对接 Gateway WebSocket,保证维护成本最低。A lightweight, locally deployable, and multi-device synchronized Clawdbot chat client. It focuses solely on connecting v…

    TypeScript 11

  5. Steward Steward Public

    一个无感感知上下文、主动推进事务、只在关键决策点打扰用户的协作 Agent。An ambient proactive agent that handles low-risk work automatically and briefs users only when their judgment is needed.

    Python 2