The cable works great, but when powered on it will pull the reset pin low by default. I have to send some obscure command using either an FTDI-specific library or low-level USB functions. A bit more info about this would have been helpful. In case anyone else is trying to figure this out, this is the command you need (from libftdi):

Pull reset low: ftdi_set_bitmode(ftdi, 0xf1, BITMODE_CBUS)
Release reset: ftdi_set_bitmode(ftdi, 0xf0, BITMODE_CBUS)

After doing this, the device works as expected. However, if you use libftdi to transmit/receive data, you should set the baud rate to four times the actual baud rate because there's something wrong with the internal baud rate calculation.

I'm actually running the device at 2Mbaud, which is twice as high as the maximum specified speed, but it works fine for me.

One small suggestion: If you used clear heat shrink instead of black, you would be able to cover the entire board with it without blocking any labels or LEDs.

UPDATE: Your board has some EMI issues. Whenever I turn the mains switch on my power supply on or off, it seems to create an EMI spike powerful enough to reset the FTDI chip on your board. The power LED turns off for a moment and the USB device disconnects and then reconnects to the computer. I suspect the cause of the problem is that you don't have enough capacitors on the 5V line. You only have 100nF after the ferrite bead, the FTDI datasheet recommends 10nF before the ferrite bead and 100nF+4.7uF after it. This is disappointing because I was planning to use this device in a noisy environment (that's why I needed isolation in the first place).