Correlating and jointly analysing streams of multi-modal data from online social network, news, and other customized sources is increasingly important in various domains such as security, urban planning, marketing, and many more. In this project we aim to build a scalable, extensible and customizable workbench which facilitates such analysis and result visualization. We make use of latest technologies such as Spark and graph management systems for managing massive scale dynamic data, and build on recent research work in the areas of AI/ML, NLP and knowledge graphs, data mining and management.
Open Project Positions
Candidates should be skilled in one or more of the following relevant technologies:
- Spark/Hadoop and distributed data processing
- stream management through Kafka and Flink
- data management systems such as MongoDB, ElasticSearch, Neo4J etc.
- AI/ML tools such as TensorFlow and Keras and their deployment over GPU clusters
Candidates should be conversant with recent trends in machine learning and/or natural language processing, and have the ability to evaluate and implement ideas from research papers with minimal supervision. Ideal candidates should have demonstrable skills –either in the form of having built large systems or having done projects.