Masters Thesis in Data & AI: How green is your prompt? Evaluating AI prompts for environmental impact (Project CODE GREEN) Veenendaal • Info Support

Every AI prompt has a hidden cost: energy. Vague or overloaded prompts force models to burn more compute and emit more COâ. In this thesis you will investigate how prompts can be rated for both efficiency and intent through a Green Score system. The goal is to create insights and methods that guide users toward clearer, more sustainable AI interactions.

ð¡Areas of Interest: Prompt engineering, sustainable AI, green software design

This Master’s thesis is part of the CODE GREEN graduation project. Within this project, we explore how software can be made more sustainable throughout its entire lifecycle — from design and development to testing and maintenance. Your assignment is one of the building blocks through which we collaboratively develop concrete solutions that save energy, reduce COâ emissions, and make software greener.

Generative AI models such as ChatGPT offer immense value in enterprise environments, but their environmental impact increases with usage. Most users are unaware that the way a prompt is phrased can significantly affect the compute and energy required to generate a response. Vague or verbose prompts often trigger longer, more resource-intensive processing, while overly broad or multi-part prompts demand more from the model.

Currently, no feedback mechanism exists to inform users how “eco-friendly” their prompts are, neither in terms of what they ask nor how they ask it. As Info Support integrates OpenAI models into its internal assistant, this creates an opportunity to make responsible AI usage part of the default user experience.

Design and implement a green score prompting system that evaluates the environmental footprint of AI prompts, based on both content and compute complexity. Integrate this system into Info Support’s AI assistant to help users write more sustainable prompts.

You will:

1. Research prompt optimization practices, energy usage of large language models, and sustainability metrics in NLP.
2. Define scoring criteria based on two axes:
   1. Intent Sustainability: Does the prompt ask for sustainable, eco-conscious content?
   2. Compute Efficiency: Is the prompt written efficiently to minimize token usage, avoid ambiguous requests, and limit response length?
3. Design a scoring model or framework combining these aspects into a single Green Score (e.g., 0–100, with labels like “Wasteful”, “Neutral”, “Green”).
4. Build a Proof of Concept by integrating the scoring system into Info Support’s AI assistant.
5. Evaluate the impact through user feedback, test cases, or expert interviews to assess whether it improves user awareness and behavior.

About Info Support

Info Support specializes in custom software, data/AI solutions, management, and training and is active in the Finance, Industry, Agriculture, Food & Retail, Mobility & Public, and Healthcare sectors. We provide solid and innovative solutions for complex and critical software issues. Our headquarters are located in Veenendaal (NL) and Mechelen (BE). At present, approximately 500 employees are employed by Info Support.

Info Support's working method is characterized by a number of core values: solidity, integrity, craftsmanship, and passion. These core values are intertwined in our work and the way we interact with each other.

To ensure that all employees are always up to date with the latest developments, Info Support has an in-house knowledge center that eagerly satisfies the hunger for more or different knowledge and skills.

B2 language proficiency in Dutch is required.