Lead Data Scientist
7+ years of building production AI systems. Specialized in LLM orchestration, evaluation, and full-stack ML platforms. Led teams and delivered systems serving 100K+ users. Available for CET timezones
Work Experience
Lead Data Scientist | Eka.care
Bangalore, IN 01/2022 - PresentHealth care company building AI tools for doctors
- Built MedAssist AI platform - fully featured LLM client with support for remote MCPs. Built and open-sourced MCP servers for providing Indian context to the LLMs - Link
- Led open-source release of KARMA-OpenMedEvalKit library for evaluating LLMs and Health AI systems in multiple healthcare datasets across modalities and tasks - Link
- Improved the relevance of ElasticSearch driven autosuggest for medication by 160% and reduced the nDCG by 55% through query breakdown
- Created flows for understanding of uncoded entities written by doctors. Improved the codification of Symptoms and Diagnosis by 30% and Medications by 80% through BERT based NER and NEL models
- Through Apache Beam, built a data aggregation engine to collate medical information across multiple data sources
- Deployed models through TorchServe, RayServe hosted on Kubernetes
- Gave multiple internal tech-talks to upskill team on usage of LLMs and AI
- Lead a team of two engineers and one data scientist
Data Scientist | Udaan
Bangalore, IN 05/2021 - 12/2022B2B marketplace
- Improved search to add-to-cart funnel conversion through a tree-based approach for search ranking by 10%. Iteratively A/B-ed the change across multiple business verticals
- Built a pipeline to estimate the volume of different shipments (ranging from clothing to bottles) at warehouse(s) through lidar based volume profilers. Initiated the data science stack for this area from the ground up. Data collection, annotation, training (DGCNN) and deployment
- Cost-savings of 40% and time reduction from 40s to 20s for inward shipments through volume profilers
Visiting Researcher - ISB Lab
Bangalore, IN 05/2020 - 03/2021National Centre for Biological Sciences (NCBS) | Academic Research lab. Balancing this role, while working full time
- Precision for Integrative Modelling (PrISM) through VAEs. An unsupervised technique to score integrative models on their precision
- Won the best poster award at NCBS - Annual Talks 2021.
Data Scientist III | SigTuple
Bangalore, IN 05/2018 - 05/2021Data science driven startup in health care
- Built retinal disease product(s) from scratch, starting with annotation strategies to deployment. Completed multiple clinical studies. Certified CE.
- Technical team lead for two products - Drishti (Fundus) & Shrava (Urine) analysis. Decided the direction of research and engineering development.
- Published 2 papers on deep learning at IEEE ISBI 2019 conference.
- Led the machine learning platform team with key contributions to the architecture. Worked with TF Serving, K8s, and cloud functions (GCF) for the inference of multiple models in a DAG. Improved TAT by 60% and reduced cost by 40%.
Publications
PrISM: precision for integrative structural models
Bioinformatics, Volume 38, Issue 15 August 2022Deep learning for weak supervision of diabetic retinopathy abnormalities
IEEE International Symposium on Biomedical Imaging (ISBI) July 2019Dynamic region proposal networks for semantic segmentation in automated glaucoma screening
IEEE International Symposium on Biomedical Imaging (ISBI) July 2019Recent Blogs
Why Model Context Protocol?
April 2025Making LLM workflows human friendly
January 2025Technical Skills
AI/ML
Retrieval
Backend
Data & Infrastructure
Education
B. Tech - Computer Science and Engineering
VIT University, Vellore, IN 2014 - 2018CGPA: 8.39/10
Achievements
Hackathon winners - AWS GenAI
August 2024Built an appointment booking agent through tool use
Impact Award - Eka.care
2023Awarded for outsized impact
Best Poster Award - NCBS Annual Talks
2021For PrISM research presentation