Neural representations
Compact, high-throughput representations for visual and physical fields.
iph.so
AI for the physical world.
iph.so is building neural systems for physical modeling: fast representations, learned operators, and simulation-aware architectures for video, fluids, materials, and complex dynamical systems.
Focus areas
Research-led systems for fields, operators, and dynamics.
Learned operators
Architectures that model how systems evolve, not just how they look.
Physical AI
Tools for AI systems that need to understand real dynamics, constraints, and interaction.
Current work
What we are building now.
- NIKA-style neural video representation
- Operator learning for fluid dynamics and Rayleigh–Bénard systems
- WebGPU demos coming soon