My best photo

I am a Staff Data Scientist with a Ph.D. in Physics. I am a seasoned data science-ML professional with a track record of building and deploying scalable ML solutions to business projects. I have experience working in automative, consumer services and ed-tech domains. I have a Ph.D. in Physics from the University of Goettingen and Max Planck Institute for Dynamics and Self-Organization where I worked on using computer vision techniques to study fracture patterns. I was also a postdoc at the International Center for Theoretical Sciences, where I used deep learning techniques to study collective behavior in biological systems. I love sharing my passion for computer vision, NLP, and soft matter. I am also very interested in the applications of ML and data science to healthcare domain. I have multiple collaborations with labs in Stanford and independent researchers with projects ranging from biostatistics to epidemiology. I ran a series of interviews on AI in healthcare. Reach out to me, and we can discuss potential scientific and industry collaborations.


August 2022-Current
Staff Data Scientist

  • Developed and deployed end to end machine learning pipelines on AWS Sagemaker for multiple machine learning models.
  • Reduced error in cost prediction by 20% using XG Boost regression models.
  • Automated workflows by building and deploying python based web applications.
  • Built reporting solutions for business KPIs using Python and R based dashboard.
  • Architected and deployed a data versioning system using data version control (DVC) across the entire team to process large datasets efficiently.

September 2021-July 2022
Senior Data Scientist

  • Designed an end to end multi-stage SageMaker model pipeline to map over 10+ million products to warranty plans
  • Eliminated manual process of mapping products to industry verticals by building a random forest classifier which achieved 85% accuracy.
  • Reduced data labeling time from order of days to hours by designing a new product taxonomy which is being used across teams.
  • Deployed and managed a Feast offline feature store for multiple ML models. Stored features for over 14+ million products.

May 2018- September 2021
Data Scientist

  • Developed an automated feedback system for student recorded video which has processed over 10,000 user videos.
  • Reduced feedback time on these videos from 2+ hours to less than 5 minutes. Deployed and managed solution on AWS ECS.
  • Built an audio feedback system to provide feedback on a user's rate of speech and filler words by generating transcripts from AWS Transcribe.
  • Wrote over 100+ Jupyter notebooks, statistics, machine learning and data analytics for training platform Refactored.ai
  • I was an integral part of the team that won MIT solve work of the future award. You can find the details here

Jan 2018- Feb 2018
Data Science Fellow
Insight Data Science

Jan 2016- Dec 2017
Postdoctoral Research Fellow
International Center for Theoretical Sciences

Dec 2011- Oct 2015
Ph.D Fellow
Max Planck Institute for Dynamics and Self organization


Nandakishore, Pawan, and Lucas Goehring. 2016. “Crack Patterns over Uneven Substrates.” Soft Matter 12: 2253–63. https://doi.org/10.1039/C5SM02389K.

Nguyen, David, Vinita Bharat, Devon Conradson, Pawan Nandakishore, and Xinnan Wang. 2021. “Miro1 Impairment in a Parkinson’s at-Risk Cohort.” Frontiers in Molecular Neuroscience 14 (August): 734273. https://doi.org/10.3389/fnmol.2021.734273.

Vinita Bharat, Roeland Vanhauwaert, Li Li, Colin M. Muir, Aarooran Sivakumaran Durairaj, Sujyoti Chandra, Yann Le Guen, Pawan Nandakishore, Chung-Han Hsieh, Stefano E. Rensi, Russ B. Altman, Michael D. Greicius, Liang Feng, Xinnan Wang “An Iron-Calcium-Miro Axis Influences Parkinsons Risk and Neurodegeneration.” bioRxiv. https://doi.org/10.1101/2022.10.30.513580.

Nandakishore, Pawan, Meng Lui, Rishab Prakash, Surya Gourneni, Rohan Sukumaran, Jonathan Berman, Swaminathan Iyer, and Christin Glorioso. 2021. “Deviations in Predicted COVID-19 Cases in the US During Early Months of 2021 Relate to Rise in b.1.526 and Its Family of Variants.” https://doi.org/10.1101/2021.12.06.21267388.