Build AWS based back-end data storage, processing and accessing application.

Build an AWS based application that collects data from IoT devices. The data is then processed - both near real time and batch - stored and inferences delivered via APIs and UI. We have a stream of data coming in from a set of IoT gateways deployed across the country. The objective is to 1. Build the server side logic on AWS to ingest this data, process it using a message aggregator (like Kafka), and store it on AWS Dynamo DB. The data itself needs to be subjected to the following: a) Low latency processing of alarm events (this will trigger a status change message to the UI application) b) Daily batch processing of events - generation of reports, push aggregated event data to pre-defined destinations, expose APIs for accessing the incremental data 2. Build a responsive web application that renders the reports (can use open source visualization tools as well) Approach and Timeline We expect this project to be completed within three weeks. The final code will need to be deployed in our AWS instance and successfully tested. We will facilitate multiple conversations with our teams to clarify any technical or project-related questions. Key data inputs from the gateway (assume as incoming JSON) 1. Device id (gateway) 2. Device id (sensor) - multiple sensors possible for a given phone number For each sensor, 3. Battery status (High, Low, None) 4. Device status (Active, inactive) 5. Event details (Leak detected, Leak not detected) 6. Time stamp Additional data inputs (entered through a UI as well as through a REST API) Geographical data associated with the house (address, phone number, number of residents, and around 50 more variables to be shared during discussion) Reports 1. Individual event history given phone number 2. Aggregated event history for a given time window (daily, weekly, monthly, YTD, custom window) 3. Any other predictive analytics that can be inferred from the data Data Data Size is approximately 250 KB per gateway every 5 minutes. There will 1000 devices sending in data to the system at peak. Deliverables The server side logic built along with a responsive UI (preferably React Native) for both low latency and reports; REST API endpoints. Should also incorporate AWS IAM. Any recommendations related to data architecture that would make this type of analysis easier in the future, and that could allow us to generate personalized insights on a production basis