“If you can not measure it, you can not improve it” (Lord Kelvin) clearly articulates the need for efficient and robust sensors in all fields of science and technology, especially in pharma, agriculture, environment, food and industrial biotechnology. However, current analytic techniques require laborious, expensive multi-step processes with highly specialized, non-portable equipment. As a solution, all living organisms have been evolving and fine-tuning specialized mechanisms to precisely monitor a vast array of different types of molecules. Reprogramming such mechanisms to produce fluorescent proteins in response of these molecules allows researchers to build Biological Sensors (BioS). Because of their biological context, BioS are cheap, fast, sustainable, portable, self-generating and highly sensitive and specific. Therefore, BioS bear the potential to become key-enabling tools to spur innovation and scientific exploration in various disciplines. However, the main bottleneck toward unlocking the full potential of BioS is the fact that all BioS development is currently fully dependent on and limited to only those biological mechanisms for which all BioS Building Blocks are fully identified and characterized. With this project, I want to expand my BioS research path and combine innovative engineering strategies with computational learning techniques to create new-to-nature BioS Building Blocks and greatly expand the portfolio of available BioS for any molecule of interest.