This project explores how AI agents can improve the sustainability and efficiency of serverless computing platforms such as AWS Lambda, Google Cloud Functions, or OpenFaaS.
Serverless computing simplifies deployment and scaling, but its growing adoption also increases energy consumption and environmental impact.
By integrating intelligent, autonomous AI agents, we aim to optimize resource allocation, deployment decisions, and code execution to achieve better energy efficiency and performance.

• Understand how serverless workflows operate and where optimization opportunities exist.
• Design and implement AI agent(s) that support sustainability-oriented decisions in serverless environments.
• Measure and evaluate the effects of AI integration on resource utilization, performance, and energy efficiency.

• Study the serverless computing model and typical workflows.
• Explore Agentic AI principles and agent design patterns.
• Define the goals and roles of the proposed agent(s).
• Determine integration points between serverless infrastructure and AI components.
• Identify key evaluation metrics (e.g., latency, energy use, cost efficiency).
• Research available tools and data sources for collecting metrics.
• Explore AI models and frameworks suitable for reasoning, decision-making, or optimization.
• Ensure all proposals are supported by research and credible sources.

• Define the architecture of the system (serverless + AI integration).
• Plan the experimental setup for testing and measuring performance.

• Build the prototype or proof of concept.
• Run experiments and evaluate results based on defined metrics.
• Document findings and discuss sustainability implications.

• Serverless Computing (FaaS): A cloud model where functions are executed on demand without managing infrastructure.
• AI Agents: Autonomous systems capable of perception, reasoning, and action to achieve defined goals.
• Sustainability in Cloud Computing: Efforts to minimize environmental impact by optimizing energy and resource consumption.

• OpenFaaS
• MCP Protocol
• LangChain
• Agentic Design Principles

ai-agents-serverless-sustainability/
├─ docs/                 # Research notes, references, diagrams
├─ design/               # Architecture drafts, experiments design
├─ src/                  # Code (agent + serverless functions)
├─ data/                 # Logs, metrics, evaluation data
├─ results/              # Experiment outputs and reports
└─ README.md             # Project overview (this file)

• tbd.

1. Clone the repository:

   git clone https://github.com/carl-egge/iosl-ai-serverless-sustainability.git
   cd iosl-ai-serverless-sustainability

2. Create a branch for your work:

   git checkout -b feature/<your-topic>

3. Add documentation or initial experiments under /docs or /src.

4. Submit a pull request when ready for review.

This project will adopt an open-source license (to be decided).

List of team members and roles will be added here as the project progresses.