Teaching and Curriculum

Teaching has always been an anchor in my work—an opportunity to translate complex ideas into shared understanding and to help learners connect data, theory, and practice. Across research and education, I aim to make learning environments transparent, inclusive, and evidence-based. My teaching integrates concepts from artificial intelligence, neuroscience, and data science with the human elements of reasoning, feedback, and curiosity.

Current Roles at Harvard Medical School

Lecturer in Biomedical Informatics · Curriculum & Research Fellow, Department of Biomedical Informatics (HMS)
Within the Department’s AI and Machine Learning in Medicine curriculum, I co-design and co-teach several graduate-level courses that bridge foundational AI methods with clinical and educational applications. My work spans course design, lecture and tutorial delivery, developing learning materials, designing assessments, and writing rubrics—all for cohorts with varied quantitative, computational, and clinical backgrounds. I also build analytic and AI-enabled tools that help instructors and students understand progress more clearly, and I integrate and evaluate those tools in authentic classroom settings.

BMI702 / AIM2 — AI in Medicine II

A Zitnik Lab course (Marinka Žitnik, Ph.D.).
AIM2 is an interdisciplinary course introducing students to the core methods of artificial intelligence in healthcare. I co-designed and co-taught course content on transformers, large language models, graph neural networks, and evaluation, alongside project-based learning in clinical AI. I built a hands-on tutorial on fine-tuning large language models—complete with working fine-tuning pipelines used as instructional examples—and contributed to interactive materials that emphasize interpretation, reproducibility, and collaboration. The full course curriculum is released as an open-source syllabus.

BMIF204 — Foundations of Clinical Data & its Applications

This active graduate course introduces students to the structure and complexity of real-world clinical data—clinical data sources, causal inference, and applied project work in biomedical informatics. I contribute to course design and delivery, helping students appreciate the nuances of data quality, bias, and generalizability as they move between data representation and clinical meaning. I also organized and managed a semester-long seminar series featuring 12 guest speakers for the course.

BMI712 — AI in Medical Imaging

BMI712 is an active course on deep learning for biomedical imaging, image analysis, and clinically oriented applications. Drawing on my Ph.D. work in medical biophysics and breast-MRI analysis, I contribute to course development and instructional materials, helping students combine imaging foundations with interpretability, clinical validation, and critical thinking around algorithmic performance and reliability in high-stakes clinical settings.

Leadership & Mentoring

• I manage 7 teaching fellows across the three graduate courses—3 in AI in Medicine II (BMI702), 3 in AI in Medical Imaging (BMI712), and 1 in Foundations of Clinical Data (BMIF204)—and train them using in-house, Canvas-based TF training modules I developed.
• I mentor student project teams end-to-end, from concept and experimental design through implementation to dissemination, across these project- and exam-based courses.

Workshops & Seminars

• I am the primary organizer of the Core of Computational Biomedicine (CCB) AI Workshop Series, a campus-facing program that runs about 4–5 workshops per semester with roughly 120 participants on average (hybrid). I lead a CCB team of around 8. Topics include prompt engineering, scientific writing and literature review with LLMs, agentic coding, and Python/R workflows in Posit.
• For BMIF204, I organized and managed a semester-long seminar series featuring 12 guest speakers, developing content and delivery logistics for audiences with varied research, teaching, and computational backgrounds.

Previous Teaching Experience

Before joining Harvard Medical School, I supported undergraduate and graduate courses across several institutions:

• University of Toronto — Teaching Assistant, Statistics (2021–2022). For STA288, Statistics and Scientific Inquiry in the Medical Sciences, I supported a large-enrollment cohort (~300–400 students) during COVID-era online delivery. I held remote office hours, guided students on term-project framing and statistical reasoning, and graded assignments and project components with written feedback aimed at improving understanding and reproducibility.
• York University — Teaching Assistant, Mathematics (2016–2017). I supported large multi-section undergraduate courses in Integral Calculus and Linear Algebra, plus a Math/Stats drop-in tutoring lab. I delivered tutorials and problem-solving sessions, assisted with grading and exam invigilation, and provided drop-in help for diverse learner needs.
• University of New Brunswick — Teaching Assistant, Engineering Statistics (2015–2016). For Introduction to Statistics for Engineering, I graded assignments and exams and held office hours supporting foundational statistical methods and applied interpretation.
• University of New Brunswick — Calculus & Statistics Tutor (2010–2014). I provided private and drop-in tutoring in calculus and statistics, focusing on concept clarity, problem decomposition, and exam preparation.

These experiences helped shape my philosophy of teaching: that clarity, curiosity, and humility are the foundations of scientific learning.

Teaching Philosophy

Across all levels, I teach with the belief that learning is a collaborative experiment. My role as an educator is to design environments where students can test ideas safely, receive meaningful feedback, and build confidence in their own analytical voice. Whether developing AI systems or teaching about them, I aim to create structures that support deep understanding and enduring curiosity.