Assess graduate-level ability to apply pruning, quantization, and knowledge distillation to compress deep neural networks for resource-constrained edge devices. Tasks include selecting and justifying appropriate pruning (structured vs. unstructured), quantization (PTQ vs. QAT, bitwidth selection), and distillation strategies; designing a hardware-aware compression and deployment pipeline; and analyzing trade-offs (accuracy, latency, memory, energy) with concrete evaluation metrics and validation experiments.