Recording neuronal activity throughout the brain with high temporal and spatial resolution may be a critical step in understanding how the brain works. Task-based approaches allow intelligent trade-offs between resolution, speed, and signal. I will describe projective two-photon imaging methods that leverage the spatiotemperal sparseness of neural activity and use holographic multiplexing and statistical source separation to create capable platforms for high performance imaging with single cell resolution. Similar holographic platforms can also be used to activate ensembles of neurons with single cell precision, and I will describe recent efforts to improve targeting and control in awake behaving animals.