In-memory Patient Centric Database is designed and developed to store Patient Records in Healthcare Domain. 
It is Distributed Database built on Golang and Bitmap Technology and deployed on Ec2 cluster. It exposes HQL (Health Query Language) and REST interface to interact with Web-Speed (milliseconds to few seconds).

It has Novel Storage Mechanism with highly compressed big data. It is massively parallel with patient partitioned. Also it is horizontal scalability.

It has built-in Access control mechanism with integrated authentication.

Designed and deployed TimeSeries StatForecast tool for Supply Chain Domain. It was developed using Classical Forecast Algorithms and Machine Learning. It has integrated Data Uploader, Data Mapper which enables customer to use it without any external support.

The forecase tool uses multiprocessing to train models for Timeseries forecasting. It has integrated Demand Planning Tool which enabled customer to review and finalise the generated forecast for Products at different hierarchy level.

Designed and Developed Cluster Management tool to manage Spot Ec2 machines for In-memory Database.

The tool is developed in Golang and has built-in capability to spin spot machines using terraform Scripts. Manages AWS Interruptions and launches spot machines with fully automated way.

It saves 80% Infra cost for the company as running Large On-demand Ec2 machines would be very costly. And this cluster manages Spot machines with zero downtime.

Designed and developed this Supply Chain tool to solve capacity planning related problems for India's Biggest Automobile Organization. The tool includes BOM (Bill of Material) flattening Algorithm and saving it in Graph Database. The tool provides integrated Frontend interface to interact in web-speed.

The computation and complex calculation is written in Python using Pandas Dataframe using Caching Layer. To speed up performance to interact Frontend with Backend layer with real time calculations, Implemented Compression using Apache Arrow & Redis. 

Designed and developed Chatbot Framework including ChatBot Widget, Channel Integration, Real Time Agent Transfer, IVR Integrated, NLP Trained model.

The chatbot can be launched in multi-tenant environment integrated with customized NLP engine. It saves user queries and responses and provide interface to Train it. The NLP engine was developed using RASA. And Solution was hosted on AWS.

It was designed in Node.JS but due to concurrency and load challenges, middleware moved to Golang.

The Chatbot Framework Middleware exposes REST Interface for Mobile, Web & IVR.