Hi! I am Saki Imai
an aspiring Software Engineer and a student researcher at Colby Collegecurrently based in the United States. 
Machine LearningAs a Machine Learning practitioner, my work revolves around utilizing advanced technologies to enhance accessibility of STEM domains for blind or visually blind students.
Web DevelopmentI design and develop websites that enhance user experiences and optimize task efficiency. Explore the showcased projects below for a glimpse of my work.
ResearchI have actively contributed to various research projects encompassing a wide spectrum of disciplines, including eye tracking studies, user experience studies, and machine learning projects.
Reactflow Course ViewerA course search interface for Colby College built using React and React Flow library. The project is based on Next.js and was bootstrapped with create-next-app.
Accessible STEM EducationOur project goal is to develop an automated STEM graph description model that communicates both the embedded text data and inherent spatial information into theory-based accessible captions.
Eye tracking researchDesigned and developed efficient PyQt5-based eye tracking analysis software, enabling a comprehensive comparison between human pair-programming and pair-programming with GitHub Copilot. Presented my findings at the ACM Student Research Competition.
ML to detect Cyber AttacksWe conducted a comparative analysis between classical machine learning algorithms and brain-inspired neural networks to detect network attacks using the UNSW-NB15 dataset. Our findings were summarized in a poster presentation
CS Research Assistant @ Davis Institute for Artificial IntelligenceLatest
Summer 2023 Accenture Work Prep Participant
Software Development Intern @ Jackson Laboratory
CS Research Assistant @ Software Engineering and Human Factors Laboratory
- Conducted and analyzed 21 eye tracking experiments using EyeLink 1000 Plus, providing insights into user behavior/subject matter.
- Developed efficient PyQt5-based eye-tracking analysis software, streamlining data analysis.
- Published and presented my paper titled “Is GitHub Copilot a Substitute for Human Pair-programming? An Empirical Study” at ICSE ’22: 44th International Conference on Software Engineering 2022 ACM Student Research Competition.
CS Research Assistant @ Immersive Navigation Systems and Inclusive Technology Ethics Lab (INSITE)
- Classified 3000+ STEM charts/diagrams with Tensorflow and Singularity and achieved 97.27% accuracy.
- Analyzed 300+ STEM diagram descriptions with NLTK, revealing valuable insights for future project development.
- Collaborated with University of Maine researchers to establish a standardized user study protocol, improving study result consistency.
Bachelor in Computer Science & Mathematics Major @ Colby College