CS student at Monash University Malaysia. I care about AI as a long-term research craft: representation, explanation, language, and the mathematical structure underneath learning systems.
I am moving away from treating software engineering as the main identity. I still build systems, but mostly because they give me better instruments for research, writing, and long-running thought.
The questions I keep returning to:
- Cross-modal representation learning: how vision, language, geometry, and 3D signals can share structure without collapsing into shallow alignment.
- Explainability: how to inspect model behavior without reducing explanation to pretty saliency maps or post-hoc stories.
- Ontological systems and world models: how models organize objects, relations, events, affordances, and latent rules of a world.
- Linguistics, mathematics, and computer science: how formal structure, semantics, optimization, and computation meet inside intelligent systems.
- ComfortStereo / FYP: comfort-aware 2D-to-3D and stereo video generation, with an emphasis on evaluation, controllability, and human viewing constraints.
The icon set here uses Shields.io badges backed by Simple Icons.
AI / data
Research writing
Systems and workflow
Web surfaces
