Romina Mahinpei

I am a fully-funded Master of Science & Engineering (MSE) student at Princeton University’s Computer Science Department.

I received a Bachelor’s of Science (BSc) degree in Honours Computer Science along with a Mathematics Minor at the University of British Columbia (UBC). During my undergraduate degree, I had the pleasure of working with the Human-AI Interaction Lab, the Scientific Computing lab, and the Systopia lab at UBC’s Computer Science Department. As part of my MSE degree, I am currently working with the Humans and Machines lab directed by Dr. Manoel Horta Ribeiro.

My experiences span human-computer interaction (HCI), artificial intelligence (AI), applied machine learning, and systems research. My current research interests are in human-centred AI, computational social science, and social computing.

Email | CV | LinkedIn | OrcID | Google Scholar

Projects

Interactive Theorem Provers (ITPs) for Proof Education [Paper] [Code]

Conducted a user study and case study analyzing how students develop proofs using the Coq ITP, the Lean ITP, and traditional methods, focusing on usability and support for CS proof techniques.
Evaluated Coq and Lean using Nielsen’s usability heuristics to identify challenges such as unclear error messages and limited support for example-driven learning.
Synthesized findings to inform improvements to ITP design and provide practical guidance for instructors considering ITPs in CS education.

An Emergent Bottom-Up Categorization of Students’ LLMs Usage in an Undergraduate Research Course [Paper]

Conducted a thematic analysis of students’ self-reported usage patterns in third-year, multidisciplinary undergraduate research methods course targeting critical thinking skills.
Developed a bottom-up categorization of how students use LLMs, grounded in actual student behaviour.

Provenance Design and Evolution in a Production ML Library [Poster] [Code] [Tutorial]

Studied the current state of data workflows across users from academia (N = 6) and industry (N = 6) and identified ways in which data provenance could simplify those workflows.
Designed, implemented, and tested the Model Card package for Tribuo, Oracle’s open-source Java ML library, to allow Tribuo users to create partially automated machine learning model documentation.

A Generalized Framework for Describing Question Randomization [Paper] [Poster] [Code] [Video]

Conducted a thematic analysis of computer-based assessment questions in two undergraduate CS courses to investigate how instructors integrate randomization into their digital assessment questions.
Developed a six-level framework to systematically categorize these randomization strategies and validated the framework by applying it to questions from an undergraduate data structures course (N = 86) and an undergraduate computer systems course (N = 100), assessing inter-rater reliability using percentage agreement and Gwet’s AC1 coefficient.

Mixed Precision Minimal Residual (MINRES) Method [Paper] [Code]

Developed two mixed precision variants of the Minimal Residual (MINRES) method for solving sparse saddle-point linear systems, allowing users to adjust the precision of preconditioner solves and matrix-vector products.
Implemented these variants and a baseline version using CUDA C and conducted a comparative performance study on NVIDIA’s GeForce RTX 3070 Ti, evaluating speed-up across three Maxwell and three Stokes saddle-point systems.

Publications

Romina Mahinpei, Manoel Horta Ribeiro, Mae Milano. 2025. Interactive Theorem Provers for Proof Education. [Paper]. In Proceedings of the 2025 ACM SIGPLAN International Symposium on SPLASH-E (SPLASH-E ‘25). DOI: 10.1145/3758317.3759679.
Ivan Orozco Vasquez †, Romina Mahinpei †, Noureddine Elouazizi, Cristina Conati. 2025. An Emergent Bottom-Up Categorization of Students’ LLMs Usage in an Undergraduate Research Course. [Paper]. Artificial Intelligence in Education (AIED 2025). Lecture Notes in Computer Science, Volume 15881. DOI: 10.1007/978-3-031-98462-4_17.
Adam Craig Pocock, Joseph Wonsil, Romina Mahinpei, Jack Sullivan, Margo Seltzer. 2025. Provenance Design and Evolution in a Production ML Library. [Poster]. Championing Open-source DEvelopment in ML Workshop @ ICML25 (CODEML @ICML 2025). OpenReview: https://openreview.net/forum?id=VrbDf3UDgv.
Romina Mahinpei †, Iris Xu †, Steven Wolfman, and Firas Moosvi. 2025. A Generalized Framework for Describing Question Randomization. [Paper]. In Proceedings of the 2025 ACM Conference on International Computing Education, Volume 1 (ICER 2025). DOI: 10.1145/3702652.3744222.
Romina Mahinpei †, Iris Xu †, Steven Wolfman, and Firas Moosvi. 2024. A Generalized Framework for Describing Question Randomization. [Poster]. In Proceedings of the 55th ACM Technical Symposium on Computer Science Education, Volume 2 (SIGCSE TS 2024). DOI: 10.1145/3626253.3635599.
Romina Mahinpei, Chen Greif. 2024. Mixed Precision MINRES. [Paper]. SIAM Undergraduate Research Online, Volume 17 (SIURO). Society for Industrial and Applied Mathematics. DOI: 10.1137/24s1678489.

† Both authors contributed equally to this work.

Teaching

As a teaching assistant, I’ve had the pleasure of assisting with the following courses:

COS 324 Introduction to Machine Learning, Princeton (2024-2025)
CPSC 313 Hardware & Operating Systems, UBC (2023-2024)
CPSC 210 Software Construction & Development, UBC (2022-2023)
Science One Differential & Integral Calculus, UBC (2021-2022)