Nov 2013 – Aug 2014 · 9 mos · Bangalore

• As a part of Content Filtering Team at Linkedin, I :
• Designed and developed low SLA architecture for capturing spam using regular expressions.
• Designed and developed content review infrastructure for filtering adult or user flagged content.
• Led the migration of content filtering library from Spring to Offspring framework.
• Apart from development, my responsibilities included :
• Interviewing candidates for LinkedIn Bangalore : Telephonic, Onsite as well as Campus Recruitment
• Mentoring interns and new college grads

Sep 2013 – Nov 2013 · 2 mos · Bengaluru, Karnataka, India

• Experiments with Machine Aided Learning Framework
• Compared Semi-Supervised Label Propagation algorithms to generate additional labeled instances and use them subsequently to rank products for search queries.
• Developed a metric to measure quality of a Product-Label graph for a query – serves a query selection factor.

Jul 2012 – May 2013 · 10 mos · Greater Lucknow Area

• Tutor for ESC101 course - Fundamentals of Computing.
• Conducted tutorial classes for first year students once every week.
• Assisted instructor in making, conducting and grading weekly quizzes, laboratory assignments and exams.

May 2012 – Jul 2012 · 2 mos · Singapore

• Semantic Role Labeling using higher Order SemiCRFs
• We compare the effectiveness of Higher Order SemiCRFs in learning Semantic Role Labeling, exploiting the parse structure of a sentence.
• We work on OntoNotes dataset because of its high inter-annotator agreement.
• We compare our systems against the CoNLL-2005 baseline and CRFs using the same features.
• We implemented a less computationally expensive and more ram intensive version of the Higher Order SemiCRFs : https://github.com/nvcuong/HOSemiCRF

May 2011 – Jul 2011 · 2 mos · Bangaluru

• Web Search Query Segmentation
• Studied the effects of incorporating query frequencies and linguistic patterns like word order and word distances in an unsupervised learnt model by Mishra et.al(2011)
• Predicted relevance of a webpage with respect to a search query using segmentation algorithm and segment to document matching at multiple granularity.
• Implemented and online demo of the segmentation model and its variants.