All the arguments about profitability aside, how would one actually go about implementing it in a sensible way? what's the compatibility list? How much memory do we actually need to calculate the hash, and how do we send/receive it efficiently?
All these questions are good ones to answer in an arduino implementation, and could have efficiency consequences for custom boards/IC's afterward.
AVR has a GNU toolchain so you'd just need to split the code above, do the hashing on an AVR that doesn't need much RAM and pass it back to a PC say over a serial port. For the reasons above anything using an AVR would be a waste of time for a custom board, and a custom IC would be an ASIC. Apart from being very expensive starting with an AVR isn't even vaguely how you'd go about starting an ASIC design. There's a thread here about an open source FPGA design which would be more likely a starting point for an ASIC:
http://forum.bitcoin.org/index.php?topic=9047.0But for an ASIC you'd want to count on dropping over $100K even if you were an expert in the area.