AI mini workshop S2026

Leveraging Generative AI in Graduate STEM Research and Education

Format: 10 weeks · 1 hour per week · hands-on

Audience: STEM graduate students (with emphasis on atmospheric physics and related fields)

Primary tools: ChatGPT and Gemini

Instructor: Dr. Zhibo Zhang (zhibo.zhang@UMBC.edut)

Time and location: Every Monday 10:30AM-11:30AM Conference room 401

Prerequisites: None. Basic familiarity with coding and academic research expected.


Course Description

Generative AI tools are rapidly transforming how researchers learn, code, write, and conduct scientific inquiry. This mini seminar series introduces graduate students to practical, ethical, and research-oriented uses of generative AI, with an emphasis on hands-on experimentation and critical evaluation. Students will learn how to use AI tools effectively for learning, coding, literature review, research design, and academic writing, while understanding their limitations and ethical implications.

Each session combines short lectures, live demonstrations, guided hands-on practice, and group discussion.


Learning Objectives

By the end of this seminar series, students will be able to:

  • Use generative AI tools effectively for graduate-level learning and research

  • Apply AI responsibly to coding, debugging, and scientific workflows

  • Leverage AI for literature review, research brainstorming, and writing support

  • Critically evaluate AI outputs for accuracy, bias, and limitations

  • Understand ethical guidelines and best practices for AI use in academia


Weekly Schedule

Week 1. Introduction to Generative AI for Graduate Research

Overview of large language models, strengths and limitations, comparison of ChatGPT and Gemini, and realistic expectations for STEM research use.

Week 2. Generative AI for Learning and Conceptual Understanding

Using AI as a tutor for complex scientific concepts, multi-level explanations, Socratic prompting, and self-assessment.

Week 3. Generative AI for Coursework and Problem Solving

Ethical learning support, step-by-step reasoning, exam preparation, and distinguishing assistance from misconduct.

Week 4. Generative AI for Coding I: Code Generation and Understanding

Prompting for scientific code, explaining unfamiliar scripts, documentation, and common failure modes.

Week 5. Generative AI for Coding II: Debugging and Optimization

Debugging strategies, performance optimization, code refactoring, and workflow integration.

Week 6. Generative AI for Literature Review and Research Mapping

Using AI to survey research landscapes, identify themes and gaps, and avoid fabricated citations.

Week 7. Generative AI for Research Design and Brainstorming

Generating and stress-testing research questions, methods, and hypotheses while avoiding confirmation bias.

Week 8. Generative AI for Writing and Publication Assistance

Improving clarity and structure, drafting abstracts and cover letters, maintaining author voice, and understanding journal policies.

Week 9. Ethical Guidelines and Responsible AI Use in Academia

Academic integrity, transparency, bias, data privacy, disclosure norms, and real-world case studies.

Week 10. Capstone: Hands-On Projects and Group Discussion

Student demonstrations of AI use cases, discussion of failures and lessons learned, and development of personal AI best-practice workflows.


Instructional Approach

  • Short lectures (20–25 minutes)

  • Live demonstrations

  • Guided hands-on exercises

  • Group discussion and reflection


Ethics and Academic Integrity

This seminar emphasizes responsible and transparent use of AI. AI tools are presented as aids for learning and research, not replacements for original thinking, analysis, or authorship. Students are expected to follow departmental and university policies regarding AI use.


Expected Outcomes

Students will leave with:

  • Practical AI workflows for research and coursework

  • Improved efficiency in coding and writing tasks

  • A critical understanding of AI limitations

  • A personal framework for ethical AI use in graduate research