Assess the ability to apply model compression techniques—quantization, structured and unstructured pruning, and knowledge distillation—to design and deploy deep neural networks on resource-constrained edge devices. Tasks include selecting appropriate compression strategies given constraints (memory, latency, energy, accuracy), outlining end-to-end compression and deployment pipelines, specifying hardware-aware choices (bit-widths, sparsity patterns, inference frameworks), estimating performance–accuracy trade-offs, and interpreting experimental metrics to justify design decisions.