raeika maroufi

"To whom it may concern,

I am pleased to write this letter of recommendation for Raeika, who worked with me as an undergraduate research assistant on a SSHRC-funded research project involving natural language processing and machine learning. Our work focused on fine-tuning a BERT model to identify sentences that cite government open data, a task that required both technical rigor and meticulous attention to linguistic detail.

Raeika played a central role in the creation of our training dataset, which involved manually identifying and tagging citation-relevant sentences across a variety of scientific documents. This was a difficult task that required a deep understanding of context and citation behaviour. Raeika’s annotations were consistently reliable and lay the foundation for a high-quality dataset essential to the success of our model.

In addition to Raeika’s data annotation work, she also developed Python code to assist with citation tracking and preprocessing tasks. She made meaningful contributions to the codebase used for fine-tuning the BERT model, demonstrating a solid grasp of both the theoretical and practical aspects of working with transformer-based models. Her code was well-organized, readable, and thoughtfully documented.

Throughout the project, Raeika showed initiative, professionalism, and a strong aptitude for research. She asked insightful questions, absorbed feedback quickly, and was unafraid to tackle complex challenges. Her ability to balance detail-oriented data work with larger algorithmic and implementation goals was especially impressive.

I have no doubt that Raeika will bring the same diligence, creativity, and technical strength to any future research or academic endeavor. She has been a valuable member of our team, and I recommend her without reservation.

Sincerely,
Michael Groenendyk
Concordia University "