Casper-Studios/ai-eng-assignment ? reverse-engineered prompt

Build me a Python app that takes recipe data and automatically creates improved versions of recipes by reading community reviews and pulling out useful tips people have actually tried. I want it to use the provided recipe data, look at reviews that mention changes, extract the specific tweak with an LLM, apply that tweak back to the original recipe, and save a new enhanced recipe with clear citation back to the review it came from.

It should be able to run one test recipe or process everything in the dataset, and write the results out as JSON files plus a summary report. Make the output easy to inspect, with the original recipe, the changes made, the reasoning, and an enhancement summary showing what was altered and why. If you need current library docs or best practices, look them up online.