Nov 2020 – Aug 2023 · 2 yrs 9 mos · Bengaluru, Karnataka, India

Jan 2018 – Nov 2020 · 2 yrs 10 mos · Bengaluru Area, India

• Qubole - MLflow
• A Machine Learning Lifecycle Platform.
• Worked on integration of MLflow with QDS which involved major challenges in design and architecture. Like inter cluster communication using nginx proxy to communicate between clusters, version upgrade and multi version support.
• Added easy to create cluster UI to create and configure MLflow compute
• UI improvements specific for customer needs
• Optimised sql query by adding rank and partitions
• Improved API response time by adding nginx caching.
• Qubole - Jupyter
• JupyterLab is the next-generation user interface for Project Jupyter offering all the familiar building blocks of the classic Jupyter Notebook (notebook, terminal, text editor, file browser, rich outputs, etc.) in a flexible and powerful user interface. JupyterLab will eventually replace the classic Jupyter Notebook.
• Worked on Jupyter lab extension to integrate it with MLflow.
• Qubole - Apache Zeppelin
• Is an open source Web-based notebook that enables data-driven, interactive data analytics and collaborative documents with SQL, Scala. Qubole notebooks use apache zeppelin for notebooks implementation. We have heavily customised zeppelin for our requirements.
• As a part of the notebooks team
• I have developed new features, fixed existing bugs and helped with implementing customer requested features.
• Worked on the zeppelin upgrade to the latest OSS version.
• Performance improvements for zeppelin notebooks with 10x improvements by optimising angular data bindings.
• Qubole - First Class Notebooks
• FCN is the QDS wrapper around our apache zeppelin implementation. which provides user capabilities like ACL and organising notebooks in user specific directories. Managing Compute Instances associated with Notebook and Other notebook functionalities.
• As a part of the team added features like
• report generation in different formats like pdf, png, html using puppeteer
• worked on bug fixes and improvements.
• Upgrade Jquery to stable major version

Feb 2015 – Jan 2018 · 2 yrs 11 mos · Bengaluru, Karnataka, India

• Unbxd Autosuggest SDK
• Javascript based SDK for ecommerce sites to show typeahead suggestions.
• Added features like top suggestions and recent suggestions. Intelligent classification for suggestion categories. Also contributed to bug fixes and simplified configuration options by adding auto configure options..
• Unbxd Search SDK
• Javascript based SDK for ecommerce sites, where users can configure template options and results will be rendered based on the configuration provided.
• Worked on improvements and bug fixes
• Unbxd Analytics SDK
• Javascript based SDK for tracking events like page clicks, search, product view, cart.. Etc. The captured data then can be used to intelligently show recommendations, improved search suggestions.
• Added new tracking events and bug fixes. And since this SDK is included in all pages including critical pages like checkout, Load time was critical. Worked on reducing SDK size by removing dependency on Jquery.
• Search and Recommendation Console
• Merchandising console to configure search relevancy, merchandising banners, recommendations, view site analytics
• Implemented complete the Look recommendation widget, which shows recommendations on the product description page.
• Upgrade console from ember js to React.js
• Added new functionalities and bug fixes
• Internal Tools
• Automate Solr Core migration from different regions in AWS.
• Built an Internal Console for view/edit configurations of a site.
• Chameleon – Application to render different ecommerce verticals fashion, electronics, grocery, ... based on hostname name
• Customer Success Team
• Worked with enterprise customers like Ashleyfurniture, Express.