CSCI 7000-09 Fall 2026: Project Proposal
Project ideas are due to be presented in person in class (ECCS 1B14) on 09/09. When you pitch your project idea, use this template:
- what is the problem you are trying to solve?
- why is the problem important to solve?
- what is the current state-of-the-practice or state-of-the-art in solving this problem, and what are its limitations?
- what is your proposed solution?
- how would you convince the professor, classmates, and the research community that your solution is successful?
SUBMISSION PROCEDURE
You need to submit a PDF file (e.g. you can use PowerPoint or a Text Document, but please convert them to PDF) addressing the points above. Submit the PDF on Canvas, before 12PM on 09/09.
PRESENTATION PROCEDURE
During class, each of you gets 2 minutes to pitch a project idea, followed by feedback/questions from the professor and classmates. Your project proposal will be preloaded with all the PDF slides on the professor's laptop.
PROJECT IDEAS
The best ideas to work on are those that solve a problem you encountered yourself while developing code or other engineering artifacts -- this is also called "Scratch your own itch". When you are building an agent to solve a problem you face in your own daily life, you know best how to describe the problem, you know when you solved the problem (the "itch" is gone).
Project ideas proposed by professor. This is not a comprehensive list, instead it provides samples of projects that are within the scope of the class:
- conduct an empirical study to find design patterns for building AI Agents
- conduct an empirical study to find how Agents are currently supporting a process in SE (e.g., code reviewing, refactoring, etc.)
- starting from a published software development benchmark, build an Agent to solve problems from that benchmark
- design a new agentic tool to support a process or task from the software development lifecycle or from another engineering field (e.g., chemical engineering)
Here is a sample of projects that students worked on in previous semester:
- Agentic Architectural Inference
- ReactRefactor Agent
- Agent-Based Cross-Repository Pull Request Review
- Agentic UI-Grounded Tool Discovery and Refactoring in IntelliJ IDEA