Aniruddha Sanyal

Worked on architecting and developing a distributed, fault tolerant data streaming and storage solution, ingesting 1 million records per minute across 3 primary data sources, improving data transformation and ingestion speed by 45% using GRPC, Kafka, Druid and Spark. Designed dashboards and created monitoring alerts for data pipelines that improved the uptime of the network by 17% month over month using Prometheus and Grafana. Designed and developed backend framework to fetch and process data from multiple data sources maintaining a maximum latency of 500ms and 99.9% uptime. Setup cloud environment for distributed data pipelines using Cloudera. Parsed, processed and stored data in 6 different data pipelines. Planned and executed upgrades for the complete Hadoop ecosystem of Aryaka’s in-house cloud. Performing day to day operations of the data dependent systems.

Helped WNS to design and develop fraud identification solutions based on machine learning techniques. Led a team of data analysts to identify the essential predictors out of hundreds of potential parameters, avoiding model overfitting / underfitting. Designed solutions for several clients and were able to achieve 10% decrease in type II errors and 7 % increase in new fraud identification. Also ensembled supervised learning techniques with unsupervised techniques like topic modeling to improve model accuracy and reduce errors.

Led the development of exploratory data analytics module for Infosys MANA. Collaborated with COEs like IOT, Blockchain and User Experience to conceptualize and develop. Designed and developed customized machine learning solutions for multiple domains.

Conducting lectures and corresponding exams. Conducting Field Visits / Industrial Visits. Conducting technical seminars and round tables. Providing guidance for projects.

Worked here as a PHP / MySQL web developer.