I’m working on an artificial-intelligence assignment that requires both designing a simple Gridworld environment and coding an agent that lives inside it. The agent doesn’t have to be fully independent; I’ll guide certain high-level decisions, but moment-to-moment behaviour should be handled by your logic so it can run partially autonomously. Here’s the flow I have in mind. First, you’ll lay out the Gridworld itself—size, obstacles, goals, rewards—and deliver clean, well-commented code so I can tweak parameters later. Then we move to the agent. I need you to craft the perception, decision-making and action loop so it can navigate, collect rewards and avoid hazards without hard-coding every move. Reinforcement-learning techniques are welcome but a rule-based approach is fine as long as it’s modular and easy to adjust. This will be based off of a report which i can send over. This and a page of requirements for said project. To keep expectations concrete, I’ll consider the job complete when I have: • A runnable Gridworld environment (Python preferred, using Pygame, Gymnasium or a lightweight custom framework). • The agent code wired into that world, demonstrating partially autonomous play for several episodes. • A short README explaining how to install dependencies, launch the simulation, and modify world or agent parameters. If you enjoy experimenting with AI behaviour and can write clear, reusable code, let’s get started—I’ll share test cases and further constraints as soon as you’re on board.