NIKHIL KINI ([email protected]) (951) 823-3852 Nikhilnkini.Wordpress.Com Github.Com/Nkini Linkedin.Com/In/Nikhil-Kini-753B7751
Total Page:16
File Type:pdf, Size:1020Kb
NIKHIL KINI ([email protected]) (951) 823-3852 nikhilnkini.wordpress.com github.com/nkini linkedin.com/in/nikhil-kini-753b7751 Main interests: ● Machine learning, Natural Language Processing, and Distributed Systems. I specialize in Entity Resolution, Probabilistic Graphical Models, and Statistical Relational Learning. Technical skills: ● Python, Java, JavaScript, the Numpy stack, scikit-learn, Maven, NLTK (Expert) ● C, C++, Matlab, PHP, Ruby, Apache Spark, Android SDK, Theano (Prior Experience) Education: ● Master’s degree in Computer Science at UC Santa Cruz, 2017. GPA: 3.47. ● Bachelor of Engineering in Information Technology, Mumbai University, 2009. Work Experience ● Over six years in research and development, of which three years in Machine Learning. Adobe Systems Inc. Data Scientist Intern (June 2016 - March 2017) - 10 months ● Visitor Stitching on Cross-Device Web Logs using Probabilistic Soft Logic ● Resolved cookies belonging to the same user, handling millions of sparsely labeled cookies ● Implemented baseline Support Vector Machine (SVM) models (in scikit-learn) ● Implemented relational classification models using Probabilistic Soft Logic ● Publication: Probabilistic Visitor Stitching on Cross-Device Web Logs, WWW 2017 University of California at Santa Cruz Graduate Student Researcher (September 2016 - March 2017) ● Research on transitivity in Hinge-loss Markov Random Fields ● Improving scalability for Entity Resolution in Probability Soft Logic Teaching Assistant for CMPS 128 Distributed Systems (Spring 2016) ● Explores fundamental as well as emerging, practical as well as theoretical topics ● Co-designed and implemented final project for the class – a distributed Key Value Store Teaching assistant for CMPS 183 Web Applications (Winter 2016) ● Course teaches Web2py (Python), Software development using Scrum. ● Responsibilities include mentoring 10 teams of 5 students each. Tata Consultancy Services ● 5 years 8 months of professional experience in development of web and mobile applications, project management, and research contributions in speech and natural language processing. L1SPAN - Level1 support with Speech and NLP Researcher - R&D - Jul '13 - Aug '15 (2 years 1 Month) ● Research in error correction techniques in ASR output using machine learning and evolutionary computing methods ● Research in HCI for usability in speech input-driven applications ● Research in Question Answering systems for classical music and constrained domains ● Development with speech processing technologies - speech recognition, speaker verification ● Design & development of Android applications 1 of 2 SSSVV: web-based learning content management system to connect 100 schools with collaborative content generation and quality assurance capabilities Developer, Project Lead - Nov '10 – June '13 (2 years 6 months) ● Developer for 1 year & project lead for 18 months ● Linux Apache MySQL PHP (LAMP) expertise, Joomla & Moodle customization, Performance and Security tuning, Software configuration management ● Patent application, System design, database design, usability ● Project lead responsibilities included Project planning and leading a team of 10 members, requirements analysis, client interfacing TCS monthly magazine microsite Developer - Sep '10 – Nov '10 (2 months) ● Javascript and CSS, Wordpress customization (PHP) Open Source Technology Group PHP expert - Apr '10 – Sep '10 (5 months) ● LAMP expertise, Training in LAMP, Zend Framework and PHP security, Linux administration Academic projects (Sample) Scaling Probabilistic Soft Logic for Entity Resolution ● Wrote a distributed implementation of the PSL allowing scaling to arbitrarily large problems. ● Improved inference time from 277 mins to 34 mins with seven workers. ● Currently working on an Apache Spark implementation of PSL. Sarcasm detection in Tweets (2016) ● Classified tweets as sarcastic based on a contrast between a positive sentiment and a negative situation. Visualizing information in text documents for better readability and understanding (2015) ● Used topic modeling with Gensim, NLTK, Stanford NLP Tools, keyword extraction, HTML5 and CSS3 to improve academic paper readability online. Distributed Object Oriented File Upkeep System (2008-09) ● Distributed file system that was an implementation of the Google File System paper implemented in Java Publications ● Probabilistic Visitor Stitching on Cross-Device Web Logs, Sungchul Kim, Nikhil Kini, Jay Pujara, Eunyee Koh, Lise Getoor. (WWW 2017, April 3–7, 2017, Perth, Australia.) ● Adapting general-purpose speech recognition engine output for domain-specific natural language question answering, C. Anantaram, Rishabh Gupta, Nikhil Kini, Sunil Kumar Kopparapu. (Workshop on Replicability and Reproducibility in Natural Language Processing: adaptive methods, resources and software at IJCAI 2015, Buenos Aires, Argentina, July 26, 2015) ● Improving ASR recognized speech output for effective natural language processing, C. Anantaram, Sunil Kumar Kopparapu, Nikhil Kini, Chiragkumar Patel. (The Ninth International Conference on Digital Society, ICDS 2015, February 22 - 27, 2015 - Lisbon, Portugal) ● TCSL at the C@merata 2014 Task: A tokenizing and parsing framework to understand queries on sheet music, Nikhil Kini (MediaEval 2014 Workshop, Barcelona, Catalunya, Spain, October 16-17, 2014) Certifications ● PHP 5.3 Zend Certified Engineer ● Trinity College London - Classical guitar - Grade 4 2 of 2 .