Scanning food for contamination is a tricky job. Since the acceptable levels are quite low, depending on the amount and geometry of the food sample, and the stochastic nature of the radioactive decay, you need a lot of "events" to be able to quantify the amount of radioactivity in your food. If the food contains considerable amounts of potassium, the natural amount of radioactive 40K may also affect your readings and deliver a positive reading that's not the result of contamination.
This means, if you want to be able to test food samples within a reasonably short time to be able to decide if it's edible or rather not, you need the highest possible sensitivity, and in order to identify the radioactive isotopes, you would want a gamma energy discrimination. In order to shield all this from external radioactivity, you'll want to house the whole sensor / sample setup within a substantial lead cabinet.
All this is pretty expensive if it's a professional system, but DIY solutions are possible. You'll be looking for a scintillator (e.g. Thallium-doped Sodium Jodide) in decent shape (this stuff is highly hygroscopic) coupled to a photo multiplier or a large surface, highly sensitive semiconductor detector (all this the bigger the better because you want the gammas to deposit all their energy inside the scintillator and not have them leave the crystal after partial deposition, and you want all the light generated in the scintillator to reach the photosensitive layer of your detector), amplifier with pulse shaper and a pulse amplitude analyzer to generate a histogram of the signal. The latter can be a PC soundcard with a freeware program, but there are other solutions with a Raspberry PI Pico or similar single board computers.
Btw, today's professional food contamination analysis systems use high-purity, cryostatically controlled germanium (bulk) detectors, providing a much better energy resolution than any scintillator, albeit at painful cost...
You can use highly sensitive geiger counters for some superficial food scans and you'll probably be able to identify "really dirty" samples, but it takes time and you hardly get any information on the isotopes that cause the contamination. But care has to be taken: There are isotopes that cannot be detected with a scintillator since their decay almost doesn't produce any gamma radiation (Polonium 210 is such an example). Yet if you're looking at the aftermath of a nuclear accident (power plant) or worse..., you are primarily looking for Iodine 131 in the first few weeks after the "event" and later on mainly for Cesium 137 and Strontium 90 which all emit some gamma radiation in their decay chain (some as bremsstrahlung as a result of their beta decay).
Small, personal Geiger Counters or Dosimeters are very useful for identifying general increased background radiation or for identifying radioactive houshold items (watches, alarm clocks, fiestaware) or minerals, but most of the inexpensive instruments cannot be considered useful for any kind of quantitative analysis since their detectors aren't energy compensated. Depending on the type of detector tube used, they can be useful for identifying the type of radiation though (alpha, beta or gamma).
It's advisable to define your requirements, do some reading and then decide what solution best suits your needs.
Good luck and all the best,
Tom