Central to all animals is the ability to interpret relevant external stimuli and react appropriately. For animals to be able to recognize and react appropriately to stimuli, neurons must first develop and wire with appropriate partners to build functioning neural circuits. The visual system has served as a model system to investigate neuronal development and target specificity, but rarely has the relationship between connectivity and downstream neuronal output been investigated in the same system. A major barrier to studying neural circuit development and investigating how neurons integrate inputs from many pre-synaptic partners is the lack of systems amenable to genetic and physical accessibility over development necessary for impactful, direct experimentation. Here, we take advantage of the genetically tractable animal Drosophila melanogaster and explore the development of a functionally well-characterized sensorimotor circuit, the convergence of visual projection neurons (VPNs) onto the dendrites of a large descending neuron called the giant fiber (GF). We find two partner VPNs, encoding different visual features that target the same GF dendrite, occupy distinct territories on the GF dendrite, in part, through sequential axon arrival during development. We find that during the initial partner matching stage, pre- and post-synaptic proteins are already present and opposed to one another. Physical occupancy is important to maintain territories, as we find the ablation of one VPN results in expanded dendritic territory occupied by the surviving partner VPN, and that this compensation enables the GF to remain responsive to ethologically relevant visual stimuli. Lastly, we develop a novel ex-vivo electrophysiology preparation and perform the first electrophysiology recording from a pupal CNS neuron. Using this preparation, we record persistent, stimulus independent neural activity in the GF in pupal stages, suggesting neural activity may be important for appropriate GF development. Our data highlight temporal mechanisms for visual feature convergence and promote the GF circuit, and the Drosophila optic glomeruli where VPN to GF connectivity resides, as an ideal developmental model for investigating complex wiring programs and plasticity in visual feature convergence.