Prompt engineering for research: A hands-on introduction workshop

While generative AI tools promise to accelerate research workflows, the outputs they return can be generic, inaccurate, or disappointing. Using AI effectively in an academic context requires a foundational understanding of how these models generate text and how a structured prompt guides that process. 

This workshop equips participants with a framework for prompting that can be
directly adapted to their own research methodologies. Working through practical research use cases, participants will construct and refine prompts, compare performance, and critically examine the limitations of the technology. The session is designed both for those new to AI tools and those looking to bring greater rigour to their existing AI use. 

Dates for 2026:

• Thursday 22 October, 10am - 12pm

• Describe, at a conceptual level, how large language models generate text and why effective prompting matters. 
• Apply a structured framework of core prompting principles to construct and refine prompts across a range of research scenarios.  
• Identify how foundational principles extend into more advanced prompting techniques. 
• Critique model outputs for hallucinations, bias, and structural limitations, distinguishing between errors that prompt engineering can mitigate and those it cannot. 
• Document and log prompts in a manner that supports transparency, reproducibility and research integrity. 

• Researchers and postdoctoral students are encouraged to engage with their peers by joining Hacky Hour to post questions and exchange ideas.

Research Data Support Services
Email: researchdata@auckland.ac.nz