This study adapts the automated patch-clamp technique introduced in Kodandaramaiah et al. (2016) to slices. The approach is visually guided (using simple computer vision algorithms); the motorized manipulator is also automatically calibrated with the camera, using a pipette detection algorithm. The paper claims a 2/3 success rate, instead of 1/3 for a human operator. The code is available in Labview and Python, which is nice, but unfortunately the Python code is not in any usable form at this moment (no documentation and very few comments). I regret that a lot of technical detail is missing from the paper, in particular details of the computer vision algorithms and of the pressure control system. This control system is different from the previous one; instead of tanks with fixed pressure, it seems to use a single pump and a pressure sensor in a clever way to produce both positive and negative pressure. The drawing on Fig. 2C is the only information I could find about the system in the paper.