Hands-on intern role developing edge AI model efficiency: implement and benchmark low-bit quantization (PTQ/QAT), optimize inference kernels, profile on-device latency, throughput and memory, and collaborate with global teams to move prototypes toward production.