I can see a lot of manufacturing difficulties:
Intricate 3D shape of the clip (4 different set of dies).
Isolation coating, with some areas untouched for electrical contact.
Proper insertion of the populated circular PCB in the clip.
PCB to clip crimp or weld for a durable contact .
Finished product testing for mechanical and electrical reliability.
I'm no expert, but the cost of an automated production line for such a product must be huge.
I think that the original plan was to use Flextronics as the custom manufacturer (CM). The CM will provide the production line and usually will require the client to pay for all custom fixturing, test jigs, etc. At the current volume, I think there is going to be a question of whether this work is even in Flex's wheelhouse. $350K is just a tiny, tiny customer for someone like Flex. There's so much overhead in getting a customer like that set up, that it's just a loser from an overhead standpoint. A Flex or a Foxconn usually won't take that kind of a client on.
Knocking out the clip on a progressive die should actually be easy with some exceptions that I'm going to let Batteroo discover. I've been to Tyco and watched them bang out contacts, and it's really impressive what they can do.
You're dead on with the coating operation. That is a major gotcha that they didn't plan on. I'm assuming it will be a sprayed on coating. That's going to be tricky because someone is going to have to hand place that spindly thing into a mask. It's going to add a lot of cost to the product.
Getting the PCB onto the clip --- well I think you're right there too. That's going to be a DFM learning experience for someone.
Aside from the SMT work, the assembly production line won't be automated, I'm fairly certain about that. They don't have nearly enough money to buy automation. This will likely be standard SMT PCBs with hand assembly to the clips. If someone was really clever, they might be able to run the whole shebang though a reflow oven with some aluminum fixtures and a bit of a presoak.