Suresh Iyengar
I am a Senior Research Developer in the Applied Sciences and Engineering group at Microsoft Research India. I mainly work on applied machine learning and accessibility related projects. Prior to this, I worked for a year at Yahoo Research India on problems related to entity matching with Dr. Rajeev Rastogi. Before this stint, I worked at Microsoft Research and Development India with the Bing team, on de-duping and ranking problems of Bing events vertical. I completed my Masters in Computer Science from Indian Institute of Technology, Bombay under Prof. S. Sudarshan. My masters thesis project was exploiting an asynchronous iterator model in databases.
Recent Projects
- Automatic synthesis of extraction wrappers for email entity extraction.
- Resource Scarce ML : Most of the iOT devices relay the sensor data to the cloud to get machine learning predictions. In this project, our aim was to make these iOT devices smarter by running ML predictions model in the device itself. This will be useful in scenarios where we cannot send the data to the cloud (latency, battery, privacy and network constraints). In this project, I have contributed to the following repositories: https://github.com/Microsoft/ELL and https://github.com/Microsoft/EdgeML.
- Accessible IDEs: In this project, we aim to make IDEs like Visual Studio more accessible. We built a Visual Studio plugin called CodeTalk which makes tasks like summarizing code, navigating within code and debugging more accessible. The code is open sourced at https://github.com/Microsoft/CodeTalk.
- Email Insights: Modern email application with features like relevance-based search, tabbed panes, autocomplete, commands from query box. Available at https://www.microsoft.com/en-us/garage/profiles/email-insights/
- Other projects include Web table queries, Fuzzy name matching and Recommending emails/documents for a meeting.
Publications
- CodeTalk: Improving Programming Environment Accessibility for Visually Impaired Developers, CHI 2018. Honorable mention award.
- InLook: Revisiting Email Search Experience, SIGIR 2016.
- Hashtag Recommendation for Enterprise Applications, CIKM 2016.
- Matching product titles using web-based enrichment, CIKM 2012.
- Active sampling for entity matching, KDD 2012.
- Exploiting Asynchronous IO using the Asynchronous Iterator Model, COMAD 2008.
In the Media
Resource scarce ML: Email Insights:- https://www.zdnet.com/product/email-insights/
- https://www.pcworld.com/article/3170146/windows/microsofts-email-insights-finally-adds-some-useful-search-smarts-to-outlook.html
- https://www.neowin.net/news/microsoft-garage-has-created-a-lightweight-email-solution-email-insights
- https://economictimes.indiatimes.com/tech/software/microsofts-codetalk-shines-the-light-for-the-visually-impaired/articleshow/63995138.cms
- https://www.eweek.com/enterprise-apps/microsoft-tackles-ide-accessibility-with-visual-studio-extension
- https://www.microsoft.com/en-us/research/blog/codetalk-rethinking-ide-accessibility/
No comments:
Post a Comment