You will never get that sort of project into a plastic DIP package, it will always have to be in a fully hermetic package, simply because you need to protect the inside carbon from oxygen, and also because to collect the electrons you will need a fairly high grade vacuum as well, plus somewhere to store the generated alpha particles (helium gas) before they rise up to the pressure that will inhibit the cell operation. So minimum will be a laser welded glass seal full size 14 or 16 pin DIP package with thin wall stainless steel top and bottom, with a small glass plate inside with the active material sputtered on it, and connected to the one power pin, with the other pins all shorted together and bonded to top and bottom covers as collector. Laser welded in a ultra high vacuum chamber so that the inner cell is as close to empty as possible of any trace of gas, and likely also baked at 600C there for a few hours as well to get rid of any adsorbed gas in the inner parts.
Vacuum chamber, laser, laser aiming devices, airlock to allow parts in and out are all likely to be custom made, and production rate will be low, based on a good long number of hours per cycle per expensive jig. Only things off the shelf will be the laser itself, and the turbomolecular pumps to pull the chamber down, though the airlocks will be similar to existing SEM sample entry ports, and your window allowing the laser in would probably be an optically pure diamond window.
To scale up just put more surface area of active material, and a bigger enclosure, but there you run into limits before you have to class it as radioactive, which is why smoke detectors, while actually containing radioactive material, are not classed as hazardous radioactive waste.
A rough and ready way to DIY ( if you have a glass forming ability and can achieve neon tube levels of evacuation at home) is to use the ready made smoke detector active emitters in a vacuum tube, wonder if there are any DIY videos of this out there.