MAE-6291 Special Topics: Generative AI for Engineering Research
Leverage AI as a powerful research accelerator—learn to use cutting-edge tools to enhance your PhD research productivity
Contact Professor
What You'll Learn
Course Overview
This course aims to explicitly teach the practical research automation systems that engineering PhD students need—literature review assistance, agentic research workflows, and AI-enhanced coding environments. It will prepare engineering PhD students to leverage generative AI as a powerful research accelerator. Unlike courses that teach how to build AI systems, this course teaches you to use AI systems to meaningfully enhance your research productivity. You'll be able to navigate the rapidly evolving landscape of AI research tools—all while understanding enough about how these systems work to use them effectively and critically. The course combines three elements: technical depth appropriate for PhD-level work, hands-on experience with research automation platforms, and project-based learning that mirrors actual research workflows. Students will learn about neural network foundations, the essence of the transformer architecture, tokenization, and the stages of large language model training and inference. They will learn about prompt engineering, vector embeddings, and retrieval augmented generation—all with a focus on applying these technologies effectively in their own research work.
Learning Objectives
By the end of this course, you will be able to:
Accelerate literature reviews
Using AI-powered search, synthesis, and analysis tools that find relevant papers, extract key findings, and identify research gaps
Design and deploy agentic AI systems
That can execute multi-step research tasks, use tools autonomously, and collaborate with you on experimental design and data analysis
Transform your coding workflow
Using AI-assisted development that goes far beyond autocomplete—including VS Code extensions, computational notebooks with AI integration, and collaborative platforms
Master research automation platforms
Including HuggingFace for models and datasets, Google AI Studio for experimentation, LangChain/LlamaIndex for workflow orchestration, and vector databases for knowledge management
Implement retrieval-augmented generation (RAG)
To ground AI outputs in your research domain, creating custom knowledge bases from literature, lab notes, and datasets
Navigate model selection
Across commercial APIs (OpenAI, Anthropic, Google) and open-source alternatives, understanding capabilities, costs, limitations, and appropriate use cases
Apply prompt engineering systematically
Using techniques like chain-of-thought reasoning, few-shot learning, role-based prompting, and structured outputs
Understand AI foundations
Including neural network basics, transformer architecture, tokenization, embeddings, attention mechanisms, and the training-inference pipeline—enough to use AI effectively without becoming an ML engineer
Evaluate AI-generated content critically
For accuracy, bias, and appropriateness in academic contexts, maintaining research integrity while leveraging AI capabilities
Build custom research tools
By combining LLMs with your domain knowledge, external data sources, and computational tools specific to your engineering discipline

This course represents an investment in research productivity that will pay dividends throughout your doctoral work and beyond. As AI capabilities expand rapidly, the researchers who can leverage these tools effectively will have substantial competitive advantages in productivity, creativity, and impact.
Why Take This Course Now?
The AI landscape is evolving rapidly, and 2025 represents an inflection point where AI tools have matured enough to be genuinely useful for research while remaining accessible enough that researchers can implement them without becoming ML engineers. The next 2-3 years will see AI integration become standard in engineering research, just as computational simulation and statistical analysis software did in previous decades. Students who develop these capabilities now will be positioned to lead research groups, secure competitive positions, and produce higher-impact work throughout their careers.
Prerequisites
Required
  • PhD student status in engineering (or permission of instructor)
  • Proficient in Python programming (ability to write scripts, use libraries, debug code)
Recommended
  • Basic familiarity with command line interfaces
  • Experience with Jupyter notebooks or similar computational environments
Not Required
No prior machine learning or AI coursework needed
Register for MAE-6291 during the Spring 2026 registration period. Questions? Contact Prof. Lorena A. Barba at [email protected]
Made with