Scribd Logo
Upload

Welcome to Scribd!

  • 29 views

    Shilpi Agrawal Resume23

    Uploaded by

    Sandeep Kulkarni
    Shilpi Agrawal has extensive experience in applied machine learning and data science. She received degrees in mathematics and computer science from IIT Kanpur. She has worked at LinkedIn developing classifiers to detect spam, misinformation, and hate speech. At Visa, she built statistical models to solve business problems in payment transactions. Her work includes publications, patents, and talks on topics like viral content detection, merchant analytics, and autonomous robotics.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Shilpi Agrawal Resume23

    Uploaded by

    Sandeep Kulkarni
    29 views1 page
    Shilpi Agrawal has extensive experience in applied machine learning and data science. She received degrees in mathematics and computer science from IIT Kanpur. She has worked at LinkedIn developing classifiers to detect spam, misinformation, and hate speech. At Visa, she built statistical models to solve business problems in payment transactions. Her work includes publications, patents, and talks on topics like viral content detection, merchant analytics, and autonomous robotics.

    Original Title

    Shilpi_Agrawal_resume23

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Shilpi Agrawal has extensive experience in applied machine learning and data science. She received degrees in mathematics and computer science from IIT Kanpur. She has worked at LinkedIn developing classifiers to detect spam, misinformation, and hate speech. At Visa, she built statistical models to solve business problems in payment transactions. Her work includes publications, patents, and talks on topics like viral content detection, merchant analytics, and autonomous robotics.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    29 views1 page

    Shilpi Agrawal Resume23

    Uploaded by

    Sandeep Kulkarni
    Shilpi Agrawal has extensive experience in applied machine learning and data science. She received degrees in mathematics and computer science from IIT Kanpur. She has worked at LinkedIn developing classifiers to detect spam, misinformation, and hate speech. At Visa, she built statistical models to solve business problems in payment transactions. Her work includes publications, patents, and talks on topics like viral content detection, merchant analytics, and autonomous robotics.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    You are on page 1of 1

    Shilpi Agrawal

    ï shilpiagrawal2711 | € shilpiagrawal.com | # shilpiag2711@gmail.com | H +1 8027230363


    Education
    Indian Institute of Technology Kanpur, B.S.and M.S. in Mathematics and Scientific Computing, Minor in Computer
    Science 2013 - 2018
    University of Waterloo, Canada, Exchange Student 2016
    Authorized to work in Canada with an open work permit
    Work Experience
    Linkedin, Applied Research Engineer, India Sep 2019 - present
    ◦ Led Trust and Safety team projects in Viral Spam detection track by spearheading model design, implementation, collabo-
    rations, and productionisation; reducing Unique Members Impacted by over 7% per week
    ◦ Designed LLM-based classifiers and Deep Neural Networks to safeguard platform against spam such as Misinformation and
    Hate Speech in various content types, reducing member and disinterest reports while increasing good UMI by 12%/week
    ◦ Conceptualised a novel solution to identify CopyPasta content and launched high-precision(+93%) classifier in close collab-
    oration with teams in the US which prevented +15k members from viewing it and increasing Desktop sessions by +0.16%
    Visa Inc., Software Engineer, India Jul 2018 - Jul 2019
    ◦ Developed novel statistical models to solve key business problems such as the Merchant Category Code shift anomaly, Pin
    Routing anomaly and Chase Merchant Services anomaly in payment transaction data which saved company’s revenue
    ◦ Built and Automated data driven analytical solution such as Data Health Check, Peer Analysis tool and dashboards providing
    detailed insights of market, spend share, opportunities, etc. for merchants; leading to 50% reduction in time efforts
    Accenture, Summer Analyst, India May 2017 - Jul 2017
    ◦ Built an end-to-end ML based solution leveraging transfer learning to forecast the expected sales prediction of apparels
    ◦ Achieved 10.6% R-squared gain on GBM model by incorporating image modality by pre-training a CNN on image data
    Pervazive Automation Solutions Pvt Ltd, Research Intern, India May 2016 - Jul 2016
    ◦ Built a novel Debt Instrument for subscriber lending rate based on credentials using MLR-generated features like LTV, etc.
    ◦ Developed semi-generative flight booking chatbot with SVM trained on Twitter data for SQL generation for response retrieval
    The Smart Cube, Research Intern, India Jun 2015 - Jul 2015
    ◦ Developed an automated risk monitoring model for rating and categorization of suppliers based on the latest news at scale
    ◦ Built an end-to-end solution crawling and scrapping news articles and rating using a probabilistic sentiment analysis model
    Projects
    Scalable Bayesian Methods for Community Detection, Prof. Piyush Rai, IIT Kanpur Spring 2017
    ◦ Proposed a modified version of stochastic gradient Markov Chain Monte Carlo (SG- MCMC) algorithm for scalable inference
    in mixed-membership stochastic blockmodels (MMSB) that models the group-structure in a network
    ◦ Achieved state-of-the-art performance in inferring overlapping communities in a network by implementing modified a-MMSB
    model to use the extra information in the form of feature vector associated with each user in the network
    EM Algorithm for Currency Classification with Mixture Models, Prof. Piyush Rai, IIT Kanpur Fall 2015
    ◦ Developed a Gaussian Mixture Model on swiss-bank data set to classify fake currency and original currency using a mixture
    of two multivariate normal distribution using the Expectation Maximization algorithm
    Nifty50 Time Series Analysis, Prof. Amit Mitra, IIT Kanpur Winter 2016
    ◦ Analyzed the daily geometric returns of NSE Nifty50 over six years, conducted tests like seasonality and Dickey-Fuller, and
    diagnosed fitted models to conclude that the data displays a volatility clustering property, best modeled by APARCH(1,1)
    Autonomous Chess Playing Robot, IIT Kanpur Summer 2014
    ◦ Built a robot employing electronics, programming, and mechanical design, which incorporates a customized chess board,
    LDR sensors, microcontroller, linear sliders, and MinMax algorithm with alpha-beta pruning to enable its AI intelligence
    Publications / Talks / Patents
    ◦ Published “Viral Content Detection at LinkedIn” blog in Linkedin Engineering Blog
    ◦ Invited Reviewer in machine learning conferences like ECIR and AISTATS
    ◦ US Patent:“Modeling User-Generated Sequences In Online Services”, Filed, Linkedin 2023
    ◦ US Patent:“Methods and System for Ingenuine CopyPasta Detection”, Filed, Linkedin 2023
    ◦ US Patent:“Methods, systems, and apparatuses for detecting merchant category code shift”, Granted, Visa Inc. 2020
    ◦ US Patent:“Methods and systems for identifying electronic transaction routing anomaly” - (Filed), Visa Inc. 2019
    ◦ Publication:“Autonomous Chess Playing Robot” in IJERT, Vol.4, Issue.03, Mar’15
    ◦ Talk:“Video Virality” at Grace Hopper Celebration India (GHCI) 2020

    You might also like