Iman Deznabi
Applied Scientist at Amazon
New York, New York, United States
Summary
👤
Senior🎓
Top SchoolIman Deznabi is an Applied Scientist at Amazon with nine years of experience and a Ph.D. in Computer Science from UMass Amherst, specializing in adaptive, personalized deep learning for time-series across healthcare, climate, and demand forecasting. Her research produced state-of-the-art models like CALM-Net for stress prediction (published in Nature Scientific Reports) and MultiWave for multi-resolution sensor interpretation, and she has applied similar techniques to in-hospital mortality prediction and fraud detection. She has moved models from research to production—building a production autoscaling predictor for Cosmos DB at Microsoft—and worked on continental-scale radar and GPS datasets, handling petabyte-scale time-series and continuous-time Bayesian models. Comfortable across academia and industry, she blends rigorous probabilistic modeling with practical deep learning and a knack for personalization that consistently improves real-world forecasting performance.
9 years of coding experience
11 years of employment as a software developer
Bachelor's degree, Information Technology, 17.48/20 (3.67/4), Bachelor's degree, Information Technology, 17.48/20 (3.67/4) at University of TabrizMaster of Science (MSc), Computer Engineering, 4/4, Master of Science (MSc), Computer Engineering, 4/4 at Bilkent UniversityDoctor of Philosophy - PhD, Computer Science, 4/4, Doctor of Philosophy - PhD, Computer Science, 4/4 at University of Massachusetts AmherstNODET(national organization of developing exceptional talents)
English, Persian, Turkish, Azerbaijani
Github Skills (15)
imputation3
sequential3
arxiv3
bidirectional2
recurrent2
kernel2
medicine2
prediction2
incremental-learning1
convolution1
diabetes1
machine-learning1
time-series1
deep-learning1
survival1
Programming languages (1)
Github contributions (5)
tastanlab/DeepKinZero
May 2019 - Feb 2021
Contributions:23 commits, 7 pushes in 1 year 9 months
ideznaby/Genetic_Artificial_Neural_Network
Jan 2017 - Oct 2019
Contributions:2 pushes, 1 branch in 2 years 8 months
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.