[Paid Research | $15 | Jan 4 Deadline] How Developers Prompt LLMs for Coding (No Skill Evaluation)

Source: Dev.to

What we study

• How developers prompt and communicate with LLMs
• Interaction patterns during coding‑related tasks

Note: This study does not evaluate code quality or programming ability.

Important clarifications

• This is not a coding test.
• There is no evaluation of your programming skills.
• You do not need to submit code for grading or review.
• We only collect prompts, conversations, and LLM outputs.

What participation involves

• Completing 5 short coding‑related tasks using your usual prompting style.
• You may use any of the following:
  • Open‑source LLMs (e.g., Code LLaMA, StarCoder, DeepSeek, etc.)
  • Commercial models (e.g., GPT, Claude, Gemini)
• Screen & audio recording during the session (research use only).
• A brief exit interview about your experience.

Duration & compensation

• Up to 60 minutes total.
• $15 after completing all study tasks and the exit interview.

Recruitment timeline

• Goal: recruit 20 participants by January 4.
• Sessions scheduled on a first‑come, first‑served basis.

Eligibility

• 18 years or older.
• Experience using LLMs for coding (any skill level welcome).

Data use & ethics

• Data will be used solely for academic research.
• Findings may be published in research venues.
• All recordings will remain confidential and used only by the research team.

Survey link

https://forms.gle/zeigEGawbAmAFZ1y6

Contact

If you have any questions, feel free to comment below or contact me at:
hsjeong1927@gmail.com