Nishant Kumar
Computer VisionAI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Senior Manager - Technology at Visa
* Senior Software Engineer at Visa starting July 2017 * Experienced Software Developer passionate about solving new, interesting and challenging problems and exploring new technologies. * Graduate Degree in Computer Science at University of Southern California (Class of 2017). * SDE Intern at Amazon.com - Summer 2016 * Around 4 years of work experience in developing REST based Web Services and web applications. Interests: Big Data,Analytics,Problem Solving & Algorithms ,Application Development,Web Services GitHub: https://github.com/nishant4498
University of Southern California
Master of Science (M.S.), Computer Science
January 1, 2015 – January 1, 2017
Cochin University of Science and Technology
Bachelor of Technology (B.Tech.), Information Technology
January 1, 2006 – January 1, 2010
Visa
Senior Engineering Manager
June 1, 2022 – Present
Visa
Software Engineer - Sr Consultant
July 1, 2017 – October 1, 2022
USC Viterbi School of Engineering
Grader Artificial Intelligence(CSCI 561)
August 1, 2016 – October 1, 2022
Los Angeles Metropolitan Area
Amazon
Software Development Engineer Intern
May 1, 2016 – August 1, 2016
Greater Seattle Area
EFI
Senior Software Engineer
January 1, 2015 – July 1, 2015
Greater Bengaluru Area
EFI
Software Engineer
July 1, 2012 – December 1, 2014
Greater Bengaluru Area
EFI
Associate Software Engineer
June 1, 2011 – June 1, 2012
Greater Bengaluru Area
Tata Consultancy Services
ASE
August 1, 2010 – June 1, 2011
Bangalore,India
StuddyBuddy
September 1, 2016 – Present
• Working on geospatial feature based project StudyBuddy where students can search the active study groups around them based on their field of interest. The RESTful web service is built using JAVA, Spring, Hibernate and PostgreSQL and deployed on AWS EC2 instance. PostGIS is used for k-NN and range queries to find the nearby groups. The client side is an Android app that displays the groups on a google map interface. A user can either create a new group, look for an active group nearby or join a group.
Bi Lingual Training and Classification
April 1, 2016 – Present
This project was implemented as the Research Project for the Natural Language Processing(NLP) course. The focused of our research was on a less explored area of NLP - training a model on text given in one language and classifying related text in another language. For the purpose of our research we worked on Marathi and Hindi News classification, where we trained our classifier on Marathi news data and then used the learning to classify news articles in Hindi.The algorithm used was Naive Bayes classifier along with Bi-lingual dictionary and our self developed stemming algorithm.Our classifier achieved an efficiency of 91%. Based on this research we could show that with the help of a good Bi-Lingual dictionary and Comparable corpora,Bi-lingual text classification can achieve efficiency closer to that of Monolingual text classification.
Map Reduce with Apache pig
March 1, 2016 – Present
* Wrote Apache Pig script,which automatically do the mapping and reducing in a cluster, to count the number of characters in 5 different text files. * Used HortonWorks Hadoop Sandbox on a VirtualBox to run the Pig script.
Spatial Data Manipulation and Visualization
March 1, 2016 – Present
* Collected the GPS co-ordinates of 9 different locations in the campus and created a KML (Keyhole Markup Language) to visualize these locations in Google Earth.Also use Quantum GIS to visualize these points in ArcGIS explorer. * Also used PostgreSQL + PostGIS to compute the Convex hull for these points and also find the 3 Nearest Neighbors from a given set of points.
Naive Bayes Classifier for hotel review
February 1, 2016 – Present
Implemented a four-way(four classes) hotel review classifier which took a hotel review as input and predicted whether it is a positive ,negative ,truthful or deceptive review. This was done by training a Naive Bayes classifier with Laplace smoothing for leaning the model from the training data.This model achieved around 99% accuracy on a given test data set.
Hidden Markov Model for Sentence Part of Speech Tagging
February 1, 2016 – Present
Implemented a Hidden Markov Model(HMM) based part of speech tagger on Catalan language.The tagger was developed using Viterbi Algorithm for HMM and backtracking. The tagger achieved an efficiency of 94% on a reference data set.
UTF 16 - UTF 8 Convertor
January 1, 2016 – Present
Knowledge about character encoding and conversion from one encoding to another is one of must have skills for every Software Developer. As a part of this project I implement the logic to convert a file containing UTF-16 string into UTF-8 format string. To achieve this each character in the input string in was converted into its binary representation and then to decimal representation. Based on the decimal range of the character, the number of bytes in UTF-8 format were calculated and the then the binary representation of the UTF-16 character was packed into corresponding UTF-8 bytes based on the definition of UTF-8 standard.
Mancala Game
September 1, 2015 – October 1, 2015
Developing an African game called Mancala (http://play-mancala.com/) using Greedy, Minimax and Alpha-Beta pruning techniques.The algorithm finds the best possible move that a player can choose considering next 10-12 moves ahead.
Uninformed Search Strategies
August 1, 2015 – September 1, 2015
Implemented a water pipeline network in a city using uninformed search techniques Uniform cost search, Depth-first search and Breadth-first search. The goal is to find the shortest path from a source city to the destination city considering several paths that can be closed for certain interval of time.
Fiery System Updater
May 1, 2015 – July 1, 2015
Developed a prototype for Fiery System Updater which updates all the Print Controllers in the field with the latest features, updates, bug fixes. This is a C++ wrapper over a C DLL that uses soap calls to fetch the latest updates from cloud and pushes the updates to the server.
Fiery Setup Wizard - Web Application
July 1, 2011 – June 1, 2012
This a Web based application using PHP , HTML , Jquery ,Apache to provide customers an easy way to set up their print controller for the first time use. The various features included server settings, date-time time zone realted settings, netrowk-ip settings and print personality settings. The backend web service is a Restful web service developed using the PHP and standard REST protocols.The web application uses this web service to display and edit various print controller settings.
Hadoop Foundations - Level 1
IBM
June 24, 2026 – Present
Cultural Fit Analysis
The candidate's project portfolio shows a strong inclination towards AI, Machine Learning, and Computer Vision, which aligns well with a 'Computer Vision' target role. The diversity of projects, from academic research to practical applications, suggests adaptability and a broad interest in the field. The progression from Software Engineer to Senior Engineering Manager at Visa indicates leadership potential and career growth, though the specific contributions in those roles are not detailed. The personal projects demonstrate initiative and self-driven learning. However, the lack of detailed descriptions for professional roles makes it difficult to fully assess cultural fit beyond technical alignment.
Soft Skills & Operational Fit
The candidate's project descriptions indicate a problem-solving mindset and an ability to work on complex algorithms. The grader role for AI suggests a foundational understanding and ability to evaluate technical concepts. However, without psychometric or English test results, a comprehensive assessment of soft skills, work attitude, stress handling, and team collaboration is not possible.