Data Scientist
-
You will join the Clinical Solutioning team at Medha AI within Narayana Health and design, develop and deploy Machine Learning, Deep Learning, Computer Vision and NLP solutions for clinical analytics and healthcare innovation. You will work with clinicians to frame problems, build and validate models, and share your work through reports and research papers.
-
We are looking for strong Python skills, hands-on experience with Machine Learning, Deep Learning and Neural Network models, strong statistics, and the ability to explain technical work to technical and non-technical stakeholders.
-
Work with clinicians and business teams to turn clinical questions into well-defined ML problems, with clear success metrics.
-
Collect, clean and process healthcare data from multiple structured and unstructured sources using SQL, APIs, OCR, text parsers and data pipelines.
-
Develop, train and evaluate Machine Learning models for use cases such as risk stratification, readmission or length-of-stay prediction, and clinical decision support.
-
Design and train Deep Learning and Neural Network architectures using TensorFlow, PyTorch and Keras, including CNNs for imaging tasks and RNNs/LSTMs for sequential clinical data.
-
Build Computer Vision solutions for medical imaging, such as image classification, segmentation, object detection and diagnostic support.
-
Apply NLP techniques to clinical documentation for entity extraction, classification, summarisation and information retrieval from notes, discharge summaries and reports.
-
Design robust validation (such as temporal splits), handle class imbalance, data leakage and drift, and be able to explain model outcomes using SHAP, regression trees or similar methods.
-
Apply statistical methods where they fit better than ML, such as hypothesis testing, survival analysis and pre-post evaluations.
-
Support deployment of models and AI/ML pipelines in production with engineering teams.
-
Contribute to research papers and conference presentations based on your own and the team's work.
-
Write clear model reports and present results to clinical and leadership audiences through presentations, dashboards and reports.
-
Review code and model.
-
2–5 years of hands-on experience in Data Science, Machine Learning or Deep Learning.
-
Strong Python with NumPy, Pandas, Scikit-learn, TensorFlow, PyTorch, Keras and OpenCV.
-
Hands-on experience building and tuning Neural Networks, including CNN and RNN/LSTM architectures.
-
Practical project experience in Computer Vision, NLP or both.
-
Strong statistics: regression, hypothesis testing, confidence intervals and experimental design.
-
Deep understanding of ML fundamentals: feature engineering, cross-validation, regularisation, bias–variance trade-off, and evaluation metrics beyond accuracy.
-
Basic understanding database querying and SQL.
-
Strong analytical, problem-solving skills.
-
Excellent data visualisation, verbal and written communication skills.
-
Exposure to medical imaging formats and tooling.
-
Research publications or conference presentations.
Education: Bachelor's or Master's in Data Science, Computer Science, Artificial Intelligence, Statistics or a related field.