CV

Experience, education, research, and selected publications.

Experience

  • Feb 2025–present
    Applied Scientist
    Amazon, Fire TV Science
    • Develop and evaluate machine learning approaches that help Fire TV customers discover relevant content.
    • Work with science, engineering, and product partners to bring research into customer experiences.
    • Focus on search, recommendations, personalization, and reliable evaluation.
  • May–Aug 2024
    Applied Scientist Intern
    Amazon, Fire TV Science
    • Researched recommendation approaches for content discovery and evaluated their effectiveness.
  • May–Aug 2023
    Deep Learning for Image and Video Processing Intern
    InterDigital Communications
    • Worked on machine learning research for image and video processing.
  • Aug 2019–Jan 2025
    Graduate Research Assistant
    Rochester Institute of Technology
    • Developed uncertainty-aware models for learning from limited data, including few-shot classification and regression.
    • Studied evidential deep learning and developed methods to improve how models learn from training examples (ICML 2023; IEEE TPAMI 2026).
    • Developed Bayesian methods for adapting vision foundation models with limited examples (NeurIPS 2024).
    • Collaborated on applications in behavioral analysis and physics-informed machine learning.

Education

  • 2025
    Ph.D. in Computing and Information Sciences
    Rochester Institute of Technology
    • Dissertation: Uncertainty-Aware Meta-Learning for Learning from Limited Data
    • Advisor: Prof. Qi Yu
  • 2017
    B.E. in Electronics and Communication Engineering
    Institute of Engineering, Pulchowk Campus, Tribhuvan University

Research areas

  • Trustworthy AI, evidential deep learning, and uncertainty quantification
  • Bayesian methods, foundation model adaptation, and few-shot learning
  • Search, recommendations, and personalization

Professional service

  • Program committee member and reviewer for AI and machine learning conferences, including NeurIPS, ICML, ICLR, CVPR, AAAI, and IJCAI.
  • Reviewer for journals in machine learning and related areas.

Selected publications

  • IEEE TPAMI 2026 — Generalized Regularized Evidential Deep Learning Models
  • NeurIPS 2024 — Bayesian Parameter Efficient Fine-Tuning of Vision Foundation Models
  • ICML 2023 — Learn to Accumulate Evidence from All Training Samples
  • AAAI 2023 — Evidential Conditional Neural Processes
  • CVPR 2022 — Multidimensional Belief Quantification for Label-Efficient Meta-Learning