Leads the development of Spotify's Machine Learning Lifecycle Platform that enables ML Practitioners to develop a deep understanding of model performance with metadata & metrics infrastructure that powers Model Evaluation, Testing, and Production Monitoring of ML Models.

Led the Data Infrastructure Group that provides a platform that enables data collection, data processing, storage, reporting, experimentation, and machine learning for engineering teams at BuzzFeed.

Led the development of Zephyr’s SaaS data management platform. The platform enables seamless integration of structured and unstructured data from various sources (public, government, vendor, and customer (e.g., CMS, PubMed, Clinical Trials, NIH). Created a knowledge graph of healthcare entities (Professionals, Hospitals, Institutions & Payers) and descriptive & predictive analytics with components like; - A collection of supervised, & unsupervised ML and NLP models that performed data extraction, parsing, cleaning, normalization, record linkage/de-duplication/fuzzy probabilistic matching, topic modeling, and named entity recognition for connecting data on entities across sources without unique identifiers. - Declarative feature engineering with configurable data retrieval, cleaning, transformation, comparison, and aggregation pipelines. Support a variety of models using Predictive Modeling Markup Language (PMML).- User interface for labeling training examples for algorithm development.- Dashboards to monitor the performance and accuracy of pipeline components.

Development of billing and CRM platform for mobile, internet, broadband, TriplePlay, VoIP providers

Development of real-time prepaid billing & CRM platform deployed to all Ericson clients.