Ira Ceka

I'm a Computer Science PhD student at Columbia University, advised by Prof. Baishakhi Ray and Prof. Gail Kaiser. I'm broadly interested in AI applications for code. Throughout my PhD, I've worked on AI for security applications, such as vulnerability detection. I've also explored code reasoning for code-related tasks—specifically, investigating the techniques that underlie the ability of large language models to perform well on tasks like bug repair. I've also explored similar ideas in the context of software engineering agents, aiming to identify which components are critical for their performance and where future research should focus to improve agent capabilities. I am a recipient of the NSF Graduate Research Fellowship (NSF GRFP).

Email  /  CV  /  Scholar  /  Github

Research

project thumbnail Understanding Software Engineering Agents Through the Lens of Traceability: An Empirical Study
I Ceka, S Pujar, I Manotas, G Kaiser, B Ray, S Ramji
arXiv, 2025
project thumbnail How Does LLM Reasoning Work for Code? A Survey and a Call to Action
I Ceka, F Qiao, A Dey, A Valecha, G Kaiser, B Ray
arXiv, 2025
project thumbnail Can LLM Prompting Serve as a Proxy for Static Analysis in Vulnerability Detection
I Ceka, S Pujar, I Manotas, G Kaiser, B Ray, S Ramji
arXiv, 2024
project thumbnail Towards causal deep learning for vulnerability detection
MM Rahman, I Ceka, C Mao, S Chakraborty, B Ray, W Le
ICSE, 2024

Honors & Awards

  • Amazon Trusted AI Challenge Selected Participant (2024)
  • National Science Foundation Graduate Research Fellowship (NSF GRFP)

Teaching

  • Programming Languages and Translators - TA, Columbia University (Fall 2023)
  • Advanced Software Engineering - TA, Columbia University (Fall 2022)
  • Data Visualization using Python - Program Instructor, MIT (Fall 2021)