It has been more than 25 years since the first indicator-based autonomous pCO₂ sensors were successfully developed and deployed on moorings. These sensors, an example of which is the Submersible Autonomous Moored Instrument (SAMI), utilize a pump and valve to refresh a pH indicator solution and run blanks. More of an in situ analyzer than a sensor, the additional complexity is necessary to achieve calibration-free performance, i.e. low or no drift for year-long deployments. Fiber optics are used to bring the light from the light source contained within a pressure housing to the optical flow cell that is at pressure. The design minimizes total flush volumes to reduce reagent consumption to 100 µl (pCO₂) per measurement. The instruments have evolved from an optical design based on a white light source (tungsten lamp) with a spectrograph for detection to light-emitting-diode sources with bandpass filters and photodiodes. The nearly identical seawater pH instrument (SAMI-pH) utilizes 50 µl per measurement providing 10,000 measurements with a 500 ml reagent reservoir. Unlike the SAMI-CO₂, which uses deionized water blanks for baseline correction, the SAMI-pH records a blank in seawater for each measurement. This presentation will focus on various technical aspects of the evolution of the sensors that should be of interest to other instrument developers. The various stages of development provide insights into the sacrifices that are made to minimize power and reagent consumption and maximize performance (precision, accuracy). The prospects for further advancements in autonomous optofluidic sensors will also be discussed.