SeeAct is designed to empower vision-language agents with a two-stage pipeline: a planning module powered by large language models generates subgoals based on observed scenes, and an execution module translates subgoals into environment-specific actions. A perception backbone extracts object and scene features from images or simulations. The modular architecture allows easy replacement of planners or perception networks and supports evaluation on AI2-THOR, Habitat, and custom environments. SeeAct accelerates research on interactive embodied AI by providing end-to-end task decomposition, grounding, and execution.

The Berkeley Pacman Projects repository offers a modular Python codebase where users build and test AI agents in a Pacman maze. It guides learners through uninformed and informed search (DFS, BFS, A*), adversarial multi-agent search (minimax, alpha-beta pruning), and reinforcement learning (Q-learning with feature extraction). Integrated graphical interfaces visualize agent behavior in real time, while built-in test cases and an autograder verify correctness. By iterating on algorithm implementations, users gain practical experience in state space exploration, heuristic design, adversarial reasoning, and reward-based learning within a unified game framework.