Generative AI & LLMs Practicum for Science

Foundations of Generative AI & LLMs for Science

• Understand how Generative AI and Large Language Models support scientific research, learning, analysis, and documentation.
• Learn key concepts such as prompts, tokens, context windows, model outputs, hallucination, reasoning, retrieval, and AI-assisted workflows.
• Explore practical uses of LLMs in biotechnology, healthcare, chemistry, materials science, environmental science, and data-driven research.

Prompt Engineering for Scientific Workflows

• Design clear and structured prompts for scientific explanations, summaries, comparisons, and research planning.
• Use role-based prompting, structured instructions, examples, constraints, and output formats for better AI responses.
• Build reusable prompt templates for literature review, hypothesis generation, protocol drafting, and scientific communication.

AI-Assisted Literature Review and Research Mapping

• Use Generative AI to summarize research articles, extract key themes, compare studies, and identify research gaps.
• Organize scientific knowledge into outlines, tables, concept maps, and structured review notes.
• Learn how to validate AI-generated literature summaries using source checking and human review.

Scientific Writing, Reports, and Documentation with LLMs

• Draft scientific abstracts, introductions, reports, explanations, lab notes, and technical summaries using AI assistance.
• Improve clarity, structure, grammar, readability, and scientific tone without changing the meaning of the content.
• Use LLMs for research proposal outlines, presentation scripts, manuscript planning, and documentation workflows.

LLMs for Scientific Data Interpretation

• Use AI to interpret tables, experimental observations, analytical summaries, and research datasets.
• Convert scientific data into readable explanations, insights, limitations, and decision-support notes.
• Understand the role of human validation when using AI for scientific data interpretation and reporting.

Retrieval-Augmented Generation for Scientific Knowledge

• Understand how Retrieval-Augmented Generation helps connect LLMs with trusted scientific documents and knowledge bases.
• Explore how RAG supports literature search, document Q&A, research summarization, and scientific decision support.
• Learn basic workflows for grounding AI responses in reliable scientific sources and reducing hallucination risks.

AI for Experiment Planning and Scientific Problem Solving

• Use Generative AI to plan experiments, prepare checklists, design workflows, and identify possible limitations.
• Generate structured research questions, hypotheses, variables, controls, and expected observations.
• Apply LLMs to simplify complex scientific concepts and support interdisciplinary problem solving.

Responsible AI, Research Ethics, and Quality Control

• Understand hallucination, bias, privacy, plagiarism, citation misuse, data sensitivity, and responsible AI practices in science.
• Learn how to fact-check AI outputs, verify scientific claims, and maintain academic integrity.
• Apply human review, source validation, and ethical documentation when using AI in scientific workflows.

Capstone: End-to-End Generative AI & LLMs Science Practicum Project

• Work on a complete science-focused AI workflow involving literature review, prompt design, data interpretation, and report creation.
• Create AI-assisted outputs such as summaries, research maps, experiment plans, documentation, and presentation material.
• Build a portfolio-ready practicum project demonstrating responsible and practical use of Generative AI and LLMs in science.