The MVP x FFF Food Computer is an AI-assisted thinking and tinkering tool for harmonizing the rhythms and culinary (& nutritional) possibilities of Food Forest Flavours (ingredients grown and harvested from food forests (FFFs)) and Alternative Proteins (Commercially available, non-animal derived high-protein food products a.k.a. Minimum Viable Proteins (MVPs)). The computer remixes ingredients based on harvest dates and user preferences, creating novel and delicious recipes that are ecologically-minded and nutritionally-complete. Instead of trying to create one perfect approach to agriculture, this artistic research project uses artificial intelligence to graft together two visions of the future that are usually pitted against each other: alternative protein production and agroforestry. But can high-tech, hyper-efficient proteins pair well with regenerative microseasonal ingredients in the kitchen? The MVP x FFF Food Computer connects these two complementary, but differently optimized, farming typologies through recipes and flavor. The current prototype is tested in the Netherlands with available data from available MVP ingredients and season products from selected Food Forests. The system is created in such a way, that new data on MVPs and food forests can be added. The tool was built to explore if and how digital tools and AI can aid in the creation of viable and desirable recipes from novel ingredient combinations (specifically those from hyper-resilient food forests and those from hyper-efficient alternative protein producers). The tool was built to generate cookable recipes that users (in our case, the Center for Genomic Gastronomy) can serve to the public via a pop-up food stand, creating a forum to reflect on and evaluate the textual and edible results of the food computer, considering both the accuracy, applicability and process by which this knowledge was created, as well as what forms of culinary knowledge are included or excluded using LLM (large language model) tools.