Content Integrity Measurement - For all Meta platforms

S3XY Sales, Service, Deliveries Forecasting. (Global)

Focused primarily to work on data engineering, application development and support for a product called Solarpulse which provides multi layer data ingestion and data aggregation to facilitate live streaming real time data for monitoring and analytics of Solar PV plants.• Developed an end to end Python-Django based customer on-boarding platform to extract data from user defined Excel and process the transformed data into MongoDB thus saving 8 hours/week of manual effort and achieving error free ecosystem.• Redesigned, developed and deployed customization feature to the core services for pushing the data to third party applications that allowed users to view the live data availability report in the application.• Redesigned deployment process into CI/CD based pipeline to accommodate seamless interaction for new features and updates using Java, Springboot, Amazon S3, MySQL, IAM• Delivered microservices as customization features to the core services for pushing the data to the third party-applications that allowed users to view the live data availability report in the application.• Thoroughly debugged and redesigned the SolarpulseAlarm module to accommodate inverter failure alarms by creating the JSON object parser lookup which drives the message queue from MQTT to Redis to User Interface using Java, SpringBoot, Maven.Later I was also a part of the Teqolytics Team which is a Predictive Analytics platform that identifies under-performing sections of an asset to keep it at its optimum level.• Implemented monitoring system using Python, MQTT, Apache Kafka, MongoDB and Apache Nifi to monitor the stream of messages coming through from MQTT broker to Kafka to MongoDB at a rate of 150,000 messages/min to improve performance bottlenecks that increased stability of application by 43 %.Technology Stack: Python,Java , SpringBoot, NodeJS, MongoDB , MySQL, PostgreSQL, Redis, Docker , Amazon S3 , Apache Kafka , MQTT , Dash , PlotlyOS : Linux , Windows

• Built a lead recommendation system to locate the nearest lead present around the salesperson's location based on specified demographic parameters such as age, occupation, marital status and education. Implemented knn based clustering to enhance the robustness of the recommender system• Achieved 85.6 percentage accuracy on rigorous tests. Presented a demo of the product within 2 months.