I investigate the feasibility of a micron-sized biosensor designed to measure neural activity. The circuitry of the biosensor is composed of graphene, some photodiodes, and a light-emitting diode. Many of these biosensors would be injected into a subject’s brain, and when powered by near-infrared light, they would glow in response to a neuron’s action potential. A model for the circuitry of these biosensors is developed and tested to predict the behavior of the biosensor. The ratio of optical output power to optical input power is greatest when the resistance of the graphene is minimized. However, the signal-to-noise ratio is better when the graphene has a higher resistance. The trade-off between these two goals is optimized by a specific resistance value that provides the greatest sensitivity of the instrument. In a test run, this model achieved a 13% energy conversion rate. This efficiency could be increased by using higher-quality components. These micro-biosensors could solve several of the current challenges facing neuro-sensing technology, including measuring the behavior of a single neuron.