Cloud Analytics: A solution that allows brick-and-mortar retailers to gain insights about customer journeys throughout their stores. The system uses IP cameras to capture anonymized metadata about customer movements, then transmits it to the cloud for analysis and visualization.Data-Processor: Service processing object trajectories to generate insights. • Collaborated with the research team to integrate filtering approaches for reducing the impact of non-person trajectories on statistics by implementing components for the data processing pipeline, verifying accuracy of the implementation and running performance evaluations. • Designed and implemented the database schema and processing components to capture customer engagement through a store. Retail-WebApp: Dashboard for retailers to visualize insights.• Designed and implemented the algorithm to identify the queue-dropout behavior using customer engagement and created API endpoints to query associated statistics. • Enhanced the heatmap visualizations by using Scalable Vector Graphics (SVG) elements and filters leading to an improved user experience when analyzing customer traffic and engagement data. Parameter-Analyzer: Tool used for analyzing and selecting parameters for the person counting algorithm. • Optimized the parameter selection process to be ~30x faster by implementing a ranking algorithm integrating results from two separate analysis.Blob-Storage-Client: Cloud agnostic data access layer. • Developed the Azure implementation of the data access layer by employing the Azure Storage SDK enabling deployment of the solution to Azure in addition to AWS. Configuration-WebApp: Application to configure new customers, sites and cameras. • Led the refactoring effort for the camera placement component by evaluating its current limitations and issues, defining use-cases and workflows by coordinating with stake-holders, and implementing backend components to support various use-cases.

Personalization platform• Contributed to the design and implementation of libraries enabling incremental learning of machine learning models on Android devices.• Developed strategy to verify accuracy of personalization algorithms implemented on mobile platforms.• Integrated database encryption and cloud storage solutions facilitating secure storage and backup of sensitive information on android devices and in the cloud.• Contributed to the development of Android apps showcasing location-aware personalization for the retail domain. IoT middleware platform• Contributed to the design and implementation of security and communication modules enabling secure peer-to-peer communication between middleware hosted devices.• Developed multiple Android apps demonstrating features and use-cases for the middleware platform.• Collaborated in conducting hackathons to evaluate usability of the middleware with developers.• Mentored master’s students at CMU for their capstone project centered around the middleware platform.

• Developed a mongoDB backed node.js web application supporting user management and authentication to host video content along with its metadata. Created a responsive frontend to support user login/registration and playback of hosted video content. • Developed a web application using meteor platform supporting low latency data synchronization across multiple web/mobile clients to demonstrate immersive TV experience.

IoT middleware platform• Contributed to the design and implementation of a reliable multicast layer and distributed network security protocols to enable secure group communication between services running on middleware hosted devices.• Developed Java and Android applications to demonstrate secure group communication of user created service groups.

Website search solution team (for a leading telecom providers website)• Developed prototypes to enable quick viewing of specific search results in an overlay without navigating to the result page.• Collaborated with team to analyze and provide solutions for duplicated search results.• Assisted in documenting various aspects of solution and system design.

• Created an on-field hardware model as a testing environment• Provided cooling solutions to increase the life cycle of the ultra-capacitors installed in windmills; two solutions were subsequently approved to be implemented for testing in the field