Skip to content
View Trenza1ore's full-sized avatar
:shipit:
Writing nonsensical code to bully compilers :suspect:
:shipit:
Writing nonsensical code to bully compilers :suspect:
20 followers · 33 following

Achievements

Achievement: YOLOAchievement: Pair ExtraordinaireAchievement: QuickdrawAchievement: StarstruckAchievement: Pull Sharkx2

Achievements

Achievement: YOLOAchievement: Pair ExtraordinaireAchievement: QuickdrawAchievement: StarstruckAchievement: Pull Sharkx2

Block or report Trenza1ore

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
Trenza1ore/README.md

👋 Hiii, Hugo here, nice to meet ya~

Personal Website: https://hugohuang.com/

OpenJiuwen - Huawei's LLM Agent Development Platform/SDK: Website & Source

Just a random research engineer at Huawei Edinburgh Research Centre who loves RL & CV. Got a Star of Edinburgh Award for my work in Jiuwen. Really like games, especially Doom & Monster Hunter.

🔧 Projects I Maintain

GitCode API Pepy Total Downloads

  • Community-maintained Python SDK / CLI tool / MCP server for GitCode, the Chinese Git platform, built with user-centric API design.
  • Now the most downloaded GitCode tool & used a lot by me and colleagues!

sb3-extra-buffers Pepy Total Downloads

  • Easy-to-use compressed RL buffers with acceptable latency.
  • Stared by maintainers of SB3 & OpenAI Gym :-)

Career Emulator

🚀 Projects I Contribute to

ViZDoom Pepy Total Downloads

  • Python type hinting support, integrated as part of CMake build pipeline.
  • In-engine game object categorization.
  • Semantic Segmentation observation wrappers.
  • Parallelize & refactor unit tests.
  • Investigate & fix bugs in Github actions & Gymnasium wrapper.

Huawei openJiuwen Pepy Total Downloads

  • In charge of Knowledge Base Retrieval, Graph Store / Memory feature.
  • Code reviewer & bug hunter for most repos within openJiuwen ecosystem.
  • Makefile for code checking & formatting.

Stable Baseline3 Pepy Total Downloads

  • Improved memory efficiency of rollout buffers.
  • Added custom rollout buffer support in sbx, jax version of SB3.

🔭 Purely Personal Projects

std-raytracer

  • A feature-rich path-tracer implemented using only C++17 standard libraries (and nlohmann's json parser).
  • Support custom models and textures, multi-threading, BVH (Bounding Volume Hierarchy) acceleration and modes for distributed rendering.
  • Path-tracing features like area lights, antialiasing sampling, aperture sampling, BRDF materials and multi-bounce path tracing.

Relax-Painter

HugoSynth

  • Undergraduate coursework, synthesizer GUI app with Matlab.
  • A feature-rich synthesizer app for cross-synthesis, granular convolution, spectrogram editing, ADSR enveloping, etc.
gitanimals lang-stat

Hugo profile views

Pinned Loading

  1. sb3-extra-buffers sb3-extra-buffers Public

    [ :suspect:👾 ] ➡️ 💾 ➡️ { 🎮🕹️ } Extra Stable-Baselines3 buffer classes. Reducing RL memory usage drastically with minimal overhead.

    Python 23

  2. GitCode-API GitCode-API Public

    简单易用的GitCode Python SDK，基于GitCode REST API接口。内置 CLI 工具，并为智能体（Agent）提供集成支持(MCP、OpenAI 和 openJiuwen 工具）。Easy to use Python SDK for the GitCode REST API. Providing builtin CLI tool, and optional integ…

    Python 6 1

  3. Farama-Foundation/ViZDoom Farama-Foundation/ViZDoom Public

    Reinforcement Learning environments based on the 1993 game Doom :godmode:

    C++ 2k 442

  4. DLR-RM/stable-baselines3 DLR-RM/stable-baselines3 Public

    PyTorch version of Stable Baselines, reliable implementations of reinforcement learning algorithms.

    Python 13.4k 2.1k

  5. std-raytracer std-raytracer Public

    A ray-traced/path-traced renderer with support for custom models & textures, built from scratch with only C++ standard library and nlohmann's json parser.

    C++ 2

  6. openJiuwen-ai/jiuwenswarm openJiuwen-ai/jiuwenswarm Public

    JiuwenSwarm is an intelligent AI Agent built on openJiuwen. It extends the powerful capabilities of large language models directly to your fingertips through various communication apps you use daily.

    Python 924 179