Hire Dharmendra T. - AI Product Manager

* Made APIs for various OR optimization problems like linear optimization, Integer optimization, Constraint optimization, Assignment, Routing, Bin packing, Network Flows and Scheduling * Designed recommendation frameworks/models for telecom products and content (media applications) * Leading projects covering classification, regression, clustering, anomaly detection, image recognition, text analysis/similarity and causal analysis. Exploring and exploiting conventional machine learning methods and neural nets, CNN, LSTM, SVD etc. * Technology landscape involves Python, Spark, SQL, various big-data technologies and dashboarding tools

Worked on following AI/ML use cases for revenue growth and network experience improvement: Telecom Customer Churn prediction, Network experience analysis, Customer segmentation, Usage prediction, Route optimization, Fuzzy text matching to identify prospective acquisition base, Fiber trial fraud analysis, Detection of anomalous behaviour, Causal impact of customer interventions Built C360 - a customer insights platform with all customer touch points with Jio (demographics, usage, recharge, network experience, app usage etc) for segmenting, generating insights and customized targeting. The key components were Data Pipeline, Machine Learning Pipeline, Presentation Layer/UI Built N360 - a network insights platform that tracks the network experience for every customer at every network cell at every hour of the day, comprising coverage and quality metrics Founder member of Community of Practice on Artificial Intelligence, Machine Learning and Data Science for Skills Enhancement, Business Problem Solving, Platform Development, Research and Product Innovation Led joint projects teams with AI/ML vendors and evaluated their solutions/platforms applying it to various customer facing/network improvement use cases Liaised with experts from India/US academia for partnerships on AI workstreams

End to end management of the data ingestion from all the data sources in one place to build a data lake making the required data transformation easy and fast for further analysis Designed the data pipelines for creating various MOATs (mother of all tables) to render a 360-degree view of the customers enabling creation of a behaviour-based customers-profile which is useful in further cohort analysis Working on various use cases of Artificial Intelligence/Machine Learning that can be implemented to help increase the recharge rate and reduce the churn rate after studying customer behaviour at every event of the telecom-service cycle Designed and implemented a predictive model to trigger alert for any network outage or coverage issues basis training dataset of post-analysis of previous network issues Geographical analysis of the existing operational/retail-network to check continuity in the Jio retail-network coverage and mapping of the left-out adjacent areas for service expansion based on the 4G network coverage availability