Prerequisites:
Familiarity with the research process and basic knowledge of Python. This workshop is specifically designed for researchers who find manual coding a bottleneck in their data analysis, simulation, or visualization workflows.
Overview:
"Vibe Coding"—a paradigm shift where researchers act as directors using natural language to guide AI "agents"—allows researches to focus on scientific questions and rely on AI to handle debugging and syntax. This workshop demonstrates how to use AI tools (like Claude and GitHub Copilot) to handle a range of complex research computing tasks.
Topics Include:
- The Researcher-Agent Loop: Moving from "writing code" to "reviewing and refining" using natural language feedback.
- Scientific Tooling with AI: Using prompts to generate code for NumPy, SciPy, Pandas, and domain-specific libraries.
- Automating Data Wrangling: Building robust pipelines that handle edge cases, missing data, and unit conversions without manual scripting.
- Verification & Rigor: Strategies for "Trust but Verify"—using AI to generate unit tests and validation checks to ensure accuracy.
Contact information:
Please email mrsmith@rice.edu if you have questions about the Data@Rice workshop series.
For more information about Python and other data courses at Rice, visit https://library.rice.edu/services/data-workshops, or contact researchdata@rice.edu.
Course Materials: