Ramita Shantharam
Associate Manager, Products & Solutions — presales and ML background
Chennai, Tamil Nadu, India
Joined April 2026
Summary
Machine learning and medical-imaging research contributor — Ramita has co-authored academic conference papers and research works focused on ultrasound-based ovarian tumour segmentation and diagnostic deep-learning pipelines, showing substantive involvement in applied medical imaging research. researchgate+2
Technical product and presales experience — Transitioned from ML/data-science internships into presales and product-and-solutions roles at Newgen, indicating a blend of technical depth and customer-facing product knowledge in enterprise software contexts. theorg+1
Practical data-science foundations from internships — Completed internships at Allied Digital Services and KLA focusing on dataset construction, trend identification, and data-science engineering tasks, providing hands-on technical experience prior to full-time product roles. theorg+1
Communication and leadership through Toastmasters — Served as Vice President of Membership in Toastmasters, reflecting active engagement in public speaking, member engagement and leadership at the club level. theorg+1
Public data-science presence — Maintains a Kaggle profile that documents personal data-science and machine-learning activity, indicating ongoing involvement in the data community outside of formal employment. kaggle
Work
Education
Projects
Writing
A novel interpretable regularized cnn with a modified xlnet transformer for segmenting and classifying the ovarian cancer
March 1, 2024Academic paper proposing an interpretable regularized CNN architecture combined with a modified XLNet transformer for segmentation and classification in ovarian cancer imaging.
Performance Analysis of Ultrasound Ovarian Tumour Segmentation Using GrabCut and FL-SNNM
May 1, 2023Conference paper analyzing ultrasound ovarian tumor segmentation techniques, comparing GrabCut and FL-SNNM approaches.
Development and validation of a deep learning pipeline to diagnose ovarian masses using ultrasound screening: a retrospective multicenter study
Conference paper describing the development and validation of a deep-learning pipeline for diagnosing ovarian masses from ultrasound imaging in a multicenter retrospective study.