Transmission electron microscopes, like other scientific instruments, are becoming increasingly complicated. Consider the I2TEM in Toulouse, a dedicated TEM for electron holography and in-situ research (HF-3300 C from Hitachi) which has a cold-field emission gun, 9 lenses, 4 apertures, 3 biprisms, 18 pivot points to align, and nearly as many elements in the corrector. Operation involves more than one hundred configurable parameters, but with approximately 10300 theoretically possible configurations, one wonders if the instrument is used to its full potential. Furthermore, appropriate microscope alignment takes between twenty minutes and an hour each day, depending on the experiment and the microscopist’s effectiveness. We propose to explore the use of artificial intelligence to automate the alignment, therefore tackling the complexity and reducing the time taken by the task.