- Machine Learning Engineer with hands-on experience in model inference, and training pipelines.- Frameworks: Sklearn, PyTorch, Tensorflow, Huggingface.- Skills: docker, Git, CI/CD, Python, Machine Learning, Deep Learning, AWS.

- Worked on the development of AI/ML framework for training and deployment of ML models for Supply Chain and Warehouse Management. - Implemented REST endpoints for online and offline prediction, offline model training.- Designed and developed performance Engineering testing frameworks for the AI/ML framework developed and deployed for different use-cases.

Work on the following data science research problems- Social media content moderation by extracting features, e.g., Entity Category, Entity Glossary, Feature Sentiment, Hate Speech Classification, Global Categorization using NLP rule-based, Machine Learning and Deep Learning based approach.- Developing descriptive and predictive analytics solution for crime modeling. Crime hotspot and crime volume prediction using multiple data sources, such as, historical crime records and GIS data.- Email compliance violation detection in Medical domain using NLP rules.

- Data science solution in recruitment domain. Used NLP, ML techniques for skillset extraction, industry classification of job description.- PoC on sensitive information detection (credit card, SSN) in customer care calls based on speech-to-text conversion and rule-based information extraction.

- Data science solution for property price prediction. Used structured and unstructured data analytics for building the price predictive model.

- Research and development of Abzooba’s core Text Analytics engine XPresso. Worked on aspect-based sentiment analysis, complaint detection, emotion classification, customer buying intent detection.- Worked on domain knowledge capturing for XPresso in varied domains (retail, hospitality, banking, insurance, telecom, airlines etc). Design and implementation of customer PoCs based on XPresso.

- Design and development of java backend - Design and optimization of sql databases