ML-Enabled Game Engine (Hades)

An exploration of game engine architecture, cross-platform Vulkan, WebGL, ImGui, JoltPhysics (used in Horizon Forbidden West), SoLoud, ECS, and C++ scripting, with particular attention to creating APIs for ML training.

GradientHound

An adaptation of BloodHound from cybersecurity, brought to model architectures and policies, to extract insights across checkpoints into the state of the gradients with architecture-aware visuals.

LaTeX++

C++ perser for math expressions, looks up each glyph's outline in a loaded OpenType font (TTF / OTF), and emits an extruded triangle mesh ready for any renderer. There is no graphics-API dependency: the library returns flat vertex/index buffers for custom rendering.

Minerva - RL Framework

Implementing diversified model architectures and training algorithms in a modular and extensible framework to be able to run experiments on RL training of models. My goal is to establish consistent metrics, visualisations, compatibility with GradientHound for debugging, and enable quick research iteration.

This project is currently private and in active development.

Vanilla MuJoCo - JoltGym

Exploring educationally whether using Jolt for multi-agent training can be as or more efficient than with MuJoCo, attempting to recreate the physics of environments like Humanoid/HalfChetah, making Gym wrappers, and training RL. Investigating if there could be any gains for the virtual world, or multi-agent interacting models.

Hades Neural Engine - C++ RL Framework

Leveraging TorchLib and custom integration layers with ImGui and Wandb to enable re-use of RL methods in other C++ projects, particularly to build and train agents that can act in 3D environments alongside human-controlled agents.