This talk is centered around building real-time and lightening responsive news search engine. From streaming data processing to system architecture, all bits and pieces will be covered. Specially in case of news articles streaming in the system, challenges and whole ecosystems of aggregation to scoring algorithm will be introduced. Realisation of plug and play micro-service architecture will be another focus of this talk so that in the end audience will see how complex but fun is building news search engine. This talk will be accompanied by live demo showing how to create one focused news feed using Cliqz News Technology.
Heeren Sharma is Software Engineer in News Team at Cliqz GmbH. He completed his Masters in Computer Science at TU Munich. During writing his Masters thesis, interest of developing data-centric products sparked in him. And, real game changer came while working at Cliqz GmbH being a fresh university graduate. Since past 3 years, developing the products and services which are more oriented towards streaming data pipelines (specially for News Search), inference engines and fast responsive systems, Heeren has developed keen interest in data engineering roles i.e. from ideation to POC to production ready systems. A problem-solver by nature, he’s particularly interested in Information Retrieval and Machine Learning. Most of the time, he likes to munch in Python and use Elasticsearch as Thor’s hammer.