About Me

“A watched loss curve never converges.” - A Wise Programmer

I’m a third year Ph.D. candidate advised by Arjun Guha in the Programming Research Laboratory at Northeastern University. My interests lie at the intersection of NLP, programming languages and systems. I focus on deploying CodeLLMs safely in software applications and training CodeLLMs for novel domains.

My work spans a range of settings — from low-resource programming languages and robotics to education, 3D modeling, and beyond. I ground my research in mechanistic interpretability: the belief that model capabilities emerge from identifiable, analyzable components within the model itself.

News

3/30: I will be giving a talk on Substance Beats Style at NAACL2025 Main conference. See you there!
3/30: We released our latest preprint PhD Knowledge Not Required .

Publications

Wu, Z., Lucchetti, F., Boruch-Gruszecki, A., Zhao, J., Anderson, C.J., Biswas, J., Cassano, F., Feldman, M. Q., & Guha, A. (2025). PhD Knowledge Not Required: A Reasoning Challenge for Large Language Models. Preprint.

Paper

Demo

Lucchetti, F., Wu, Z., Feldman, M. Q., Guha, A., & Anderson, C. J.(2024). Substance Beats Style: Why Beginning Students Fail to Code with LLMs. NAACL2025 Main Conference Oral. Paper

Lucchetti, F., & Guha, A. (2024). Understanding How CodeLLMs (Mis)Predict Types with Activation Steering. Preprint. Paper

Hu, Z., Lucchetti, F., Schlesinger, C., Saxena, Y., Freeman, A., Modak, S., ... & Biswas, J. (2024). Deploying and Evaluating LLMs to Program Service Mobile Robots. IEEE Robotics and Automation Letters.

Paper

Demo

Cassano, F., Gouwar, J., Lucchetti, F., Schlesinger, C., Freeman, A., Anderson, C. J., ... & Guha, A. (2024). Knowledge transfer from high-resource to low-resource programming languages for code llms. Proceedings of the ACM on Programming Languages, 8(OOPSLA2), 677-708. Paper