I would trust software to monitor the temperature of a fish tank and sound an alarm if it goes out of limits, i guess you already have a thermostat regulating the heater on your fish tank anyway.
It would also be interesting to experiment with chemical parameters, like pH, ammonia, nitrite, nitrate, turbidity etc.., but only in an empty tank.
Just measuring pH and adjusting it would be tricky. pH sensors also need regular calibration with buffers, and biofilms accumulate on the surface of the probes. You would certainly learn a heck of a lot just building a system to sense and adjust an empty tank, for temperature and pH.