Skip to content
View jzh001's full-sized avatar

Highlights

  • Pro

Block or report jzh001

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
jzh001/README.md

Zhiheng Jiang

Computer Science & Engineering @ UCLA (Class of 2028)

LinkedIn GitHub Email


About Me

I’m an undergraduate researcher passionate about Machine Learning, Graph Neural Networks, Reinforcement Learning, and Large Language Models.
My work spans multi-objective reinforcement learning, agentic multimodal workflows, and open-source tools bridging research with practice.

  • Former Visiting Academic at USC’s Institute for Creative Technologies (U.S. DEVCOM Army Research Lab)
  • Undergraduate Researcher at UCLA Structures-Computer Interaction Lab (LLM-guided physics simulations & multi-agent systems)
  • First-author publication in Springer Social Network Analysis and Mining, cited by Spotify Research (2024)
  • 2025 Samueli Undergraduate Scholarship Recipient and Dean’s Honors List

Skills

Languages Frameworks & Libraries Areas of Focus
Python · C++ · MATLAB · Java · JavaScript · SQL · LaTeX · Bash PyTorch · HuggingFace · LangChain · LlamaIndex · ReactJS · FastAPI · Supabase Graph Neural Networks · Reinforcement Learning · Large Language Models · Computer Vision

Highlighted Work

Multi-Objective Reinforcement Learning Project (under review)
Developed a large-scale multi-objective RL system with novel evaluation metrics and Graph Neural Network (GNN) + Proximal Policy Optimization (PPO) training pipeline. Presented at SoCal CS REU Symposium 2025 at Harvey Mudd College.

Multi-Agent LLM Robotics
Designed agentic pipelines for physics-informed Neural Architecture Search and digital twin simulations, integrating OpenGL physics engines, Intel RealSense video input, and Sawyer Robots with ROS.


Publications and Talks

  • Under review (Multi-Objective Reinforcement Learning Project) – presented at SoCal CS REU Symposium 2025
  • LLM-Guided Model Development of Elastic Structures – UCLA Undergraduate Research Week 2025
  • Unveiling Music Genre Structure Through Common-Interest Communities – Springer Social Network Analysis and Mining (2022)

Pinned Loading

  1. GraphAllocBench GraphAllocBench Public

    A robust and flexible benchmark for Preference-Driven Multi-Objective Reinforcement Learning (PD-MORL).

    Jupyter Notebook 4

  2. mlx-chat mlx-chat Public

    A lightweight, professional chat UI for running mlx-community vision-language models on Apple Silicon via mlx_vlm. Generates tokens significantly faster than Ollama or LM Studio on the same hardwar…

    JavaScript 1

  3. FreeClawRouter FreeClawRouter Public

    Free-tier LLM router built on top of OpenClaw.

    Python

  4. til-ai-2024 til-ai-2024 Public

    Winning team notebooks during TIL-AI Advanced Category Competition 2024 comprising Vision Language Models, NLP Question Answering and Speech to Text

    Jupyter Notebook 5 1

  5. Unveiling-music-genre-structure-through-common-interest-communities Unveiling-music-genre-structure-through-common-interest-communities Public

    Community detection and natural language feature extraction with metal music genres

    Jupyter Notebook

  6. When2Fly When2Fly Public

    Full Stack Web Development Final Project for CS 35L Software Construction

    JavaScript 1 1