The AI Data Viz Agent leverages large language models to interpret freeform textual instructions, and then orchestrates data processing and plotting libraries to generate code and render visualizations. Users can load datasets in CSV or JSON formats or pass pandas DataFrames directly. Once a dataset is provided, the agent can filter, group, and aggregate data based on prompts like “Show average sales by quarter for top products.” It supports multiple chart types including bar, line, scatter, pie, histogram, and heatmap, with customizable colors, labels, and annotations. The agent runs as a Python package, with optional CLI and API interfaces, enabling integration into notebooks, web services, or automated reporting pipelines.

The Mesa Predator-Prey Model is an open-source, Python-based implementation of the classic Lotka-Volterra predator-prey system, built atop the Mesa agent-based modeling framework. It simulates individual predator and prey agents moving and interacting on a grid where prey reproduce and predators hunt for food to survive. Users can configure initial populations, reproduction probabilities, energy consumption, and other environmental parameters through a web-based interface. The simulation provides real-time visualizations, including heatmaps and population curves, and logs data for post-run analysis. Researchers, educators, and students can extend the model by customizing agent behaviors, adding new species, or integrating complex ecological rules. The project is designed for ease of use, rapid prototyping, and educational demonstrations of emergent ecological dynamics.