In the last decades, the artificial communication with the brain using peripheral nerve stimulation has been extensively explored in people with sensory-motor impairments. The use of innovative neural interfaces enables the interaction of bionic devices with the human somatosensory system. It has been shown that direct nerve stimulation can effectively restore somatosensation, controlling its properties by modulating the injected charge. However, efforts are still necessary to identify encoding strategies converting sensory information into neural stimulation patterns enabling intuitive and fully natural sensations. To this end, we designed, implemented, and tested biomimetic neurostimulation paradigms able “to write” synthetic sensory information into the human peripheral nervous system. Firstly, we started designing neural modulation strategies based on the indications of a realistic in-silico model emulating the natural touch coding. Then, we measured the effect of these model-based biomimetic approaches and compared them to traditional linear neuromodulations, as well as to natural touch. We performed both animal and human experimentations to measure the effects. We recorded and analyzed the natural patterns in the somatosensory neuroaxis (sural nerves, dorsal root ganglia, and spinal cord) that resulted from the electrically- and mechanically-induced stimuli in de-cerebrated cats. We observed multiple differences in neural activation comparing biomimetic neurostimulations and those traditionally used in neuroprosthetics. After identifying the optimal biomimetic patterns evoking natural sensations, we implemented them into bionic devices, and we tested the effects in patients with limb amputation. When exploited in ecological tasks, the biomimetic neurostimulations yielded higher mobility and decreased mental effort compared to traditional approaches. These findings highlight the importance of developing neuroscience-driven technology inspired by human natural systems (biomimicry). This will allow us to finally connect humans and machines, thereby improving the efficacy of neurotechnology for individuals with sensory-motor deficits.