[Research Study] How Developers Use LLMs for Code Generation

We are recruiting developers for a research study examining how real-world developers use Large Language Models (LLMs) to generate code.

What Participation Involves:

• Completing 5 coding tasks using your usual prompting style and preferred LLMs (open-source models only)

• You can use an open-source model similar to the one you use at work, as long as it doesn’t contain any company-specific fine-tuning and can be run safely on your own environment.

• Providing your prompts, interactions, and LLM-generated code for research purposes

• Screen and audio recording during the study session

• Completing a brief exit interview regarding your experience with LLM-generated code

Data Use:

• Collected data will be used to analyze developer prompt patterns

• Research findings may be published in academic venues

• All recordings will be used solely for research and will not be shared outside the research team

Duration & Compensation:

• Up to 60 minutes total

• $15 compensation after completing all study tasks and the exit interview

Eligibility & Voluntary Participation:

• Participants must be 18 years of age or older

• Participants must successfully complete the screening questions included in the survey

Survey Link:
Click here to participate

Contact
If you have any questions, please email me at [e-mail]