Pramit Kumar

Mixture of Experts (MoE) architectures have emerged as a dominant paradigm for scaling Large Language Models (LLMs). However, MoE inference on conventional hardware is constrained by three fundamental bottlenecks. These encompass the massive memory bandwidth required to fetch non-contiguous expert weights, the non-deterministic scatter-gather traffic generated by input-dependent token routing, and the tail-latency dependency imposed by synchronous expert output aggregation.

The Covid-19 pandemic has changed the way humans interact with their environment. Common touch surfaces such as elevator switches and ATM switches are hazardous to touch as they are used by countless people every day, increasing the chance of getting infected. So, a need for touch-less interaction with machines arises.