Machine Learning, Data Mining, Natural Language Processing, Information Retrieval and applications on the Semantic Web.
Stanford: Courses taken at Stanford
BITS:
Languages: Scala, JAVA, Python, C/C++, Perl, SQL.
Operating Systems: Linux, Mac, Windows.
Others: Scalding, Lucene, Pig, Hive, Matlab, LATEX!
Worked on Hadoop, Amazon's EC2, S3, Elastic MapReduce and Google's App Engine
Detection of time-pressure induced stress in speech via acoustic indicators.
Special Interest Group on Discourse and Dialogue - SIGDIAL ’10 (PDF)
Matthew Frampton, Sandeep Sripada, Ricardo Bion and Stanley Peters.
Summarization based on document probability distributions.
23rd Pacific Asia Conference on Language, Information and Computation - PACLIC ’09 (PDF)
Sandeep Sripada and Jagadeesh Jagarlamudi.
Frec
Founding Engineer (Sep 21 - Present)
Working on all aspects of building Frec!
Twitter
Software Engineer (Oct 13 - Sep 21)
@Ads: Worked on building tweet models for detecting policy violation, risk prediction.
@Ads: Worked on Brand Syndication building targeting products for our brand advertisers
to reach audiences on-twitter & off-twitter by leveraging Mopub.
@Ads: Tech lead building of Video and App install ad products.
@Content: Worked briefly to formulate plan to unify various projects aimed at increasing content generated on Twitter.
Apple Inc., Cupertino
Software Engineer (Jul 11 - Oct 13)
@Maps Data Insights: Worked on analyzing location data to improve maps.
@Internet Services Advanced Data Analytics: As part of the Internet Services Advanced Analytics team, I was involved in designing a solution for fraud detection in user reviews. I also worked on building a system to automatically check content quality in iBooks.
Stanford University
Teaching Assistant (Apr 11 - Jun 11)
Worked with Pandu Nayak and Prabhakar Raghavan on a graduate level course (CS 276 - Information Retrieval & Web Search).
BMIR, Stanford
Research Assistant (Sep 10 - Mar 11)
Involved in the 'Data Driven Medicine' project at Shah lab (Biomedical Informatics Research). Worked on large datasets to apply Machine Learning and Data Mining techniques to predict outcomes, related drugs and diseases, off-label usage.
Apple Inc., Cupertino
Research Intern (Jun 10 - Dec 10)
I was involved in designing a prototype solution for fraud detection in user reviews. I also worked on improving components in the system that detects signup fraud. (Intern work was also selected for VP presentations)
CSLI, Stanford
Research Assistant (Oct 09 - Jun 10)
Involved in the 'Dialogue Systems' project at the Center for the Study of Language and Information (CSLI). Currently looking at how audio analysis can help in identifying stress. Also, the application of these stress measures to unit segmentation is being analyzed along with appropriate actions that need to be taken once stress is detected. (Paper accepted at SIGDIAL 2010.)
Adobe Systems Inc., India
Member Technical Staff (Jul 07 - Aug 09)
Worked with the Adobe Acrobat Connect Pro team on both back-end and client-side implementation. 7.0 Release: Contributions include server XML APIs, Integration module with BlackBoard(Won the SPOT Award), 7.5 Release: Contributions include design of new DB schema, re-architecture of old schema, implementation (JAVA), web application UI (Flex) for 3rd party Telephony Adapter integration with Adobe Connect Pro.
Microsoft Research Lab, India
Research Intern (Jan 07 - Jun 07)
The project generated extraction summaries based on the hypothesis that a summary would be able to replace or act as a substitute for the document if its probability distribution is similar to that of the original document. Two new summary generation approaches were designed (a) Summary generation by extraction of sentences based on its coverage, (b) Minimum KLD Summary Generation Method and analyzed on the DUC datasets. (Paper accepted at PACLIC 2009)
Available upon request. Contact details.