Shreyas Raghavan

Shreyas Raghavan

Bengaluru, Karnataka, India
5K followers 500+ connections

Join to follow

About

PayPal’s team benefits from advanced machine learning tools designed to enhance credit…

Activity

Join now to see all activity

Experience

  • PayPal Graphic

    PayPal

    Bengaluru, Karnataka, India

  • -

    Bengaluru, Karnataka, India

  • -

    Chennai, Tamil Nadu, India

  • -

    Chennai, Tamil Nadu, India

  • -

    Chennai, Tamil Nadu, India

  • -

    Chennai, Tamil Nadu, India

  • -

    Chennai, Tamil Nadu, India

  • -

    Chennai, Tamil Nadu, India

  • -

    Chennai , India

  • -

    Chennai, Tamil Nadu, India

Education

  • Birla Institute of Technology and Science, Pilani Graphic

    Birla Institute of Technology and Science, Pilani

    -

    Birla Institute of Technology & Science, Pilani is a private institute of higher education and a deemed university under Section 3 of the UGC Act 1956. The institute is one of the first six institutes to be awarded the Institute of Eminence status in 2018.

    Program has the same curriculum as of the Full time courses , with classes over the weekend.

    The program is one of the most comprehensive masters degree programme in Data Science & Engineering without taking a break from your…

    Birla Institute of Technology & Science, Pilani is a private institute of higher education and a deemed university under Section 3 of the UGC Act 1956. The institute is one of the first six institutes to be awarded the Institute of Eminence status in 2018.

    Program has the same curriculum as of the Full time courses , with classes over the weekend.

    The program is one of the most comprehensive masters degree programme in Data Science & Engineering without taking a break from your career.
    M.Tech. Data Science and Engineering is a four-semester programme designed for working professionals that helps learners build mathematical and engineering skills required to advance their career as a Data Scientist or Data Engineer.

  • -

    SRM Institute of Science and Technology or SRMIST (formerly also known as SRM University) is a top-ranked deemed-to-be university that is part of the SRM Group of Institutions. Accredited by the NAAC with the highest 'A++' Grade, SRMIST is a member of the AIU and is approved by UGC.

    During the 4 year stint at SRM , have focussed upon the development from foundational computing to Machine Learning. Have also worked on different projects for the college which is as follows :-

    1…

    SRM Institute of Science and Technology or SRMIST (formerly also known as SRM University) is a top-ranked deemed-to-be university that is part of the SRM Group of Institutions. Accredited by the NAAC with the highest 'A++' Grade, SRMIST is a member of the AIU and is approved by UGC.

    During the 4 year stint at SRM , have focussed upon the development from foundational computing to Machine Learning. Have also worked on different projects for the college which is as follows :-

    1. PHP based website to allow requests and responses.
    2. Music Recomendation System using Python.
    3. Website made using Bootstrap , PHP , HTML and parsing data from other feeds to show statistics of English Premier League.
    4. Self Organizing Map to predict Credit fraud.

  • -

    In this program, Learning was around the techniques that power self-driving cars across the full stack of a vehicle’s autonomous capabilities. Using Deep Learning with radar and lidar sensor fusion, you will train the vehicle to detect and identify its surroundings to inform navigation.

  • -

    Activities and Societies: Debate , Prefect , Football Team

    I was the discipline incharge in my school and have also represented school in various zonal competition for Football and debate

  • -

    Activities and Societies: Extempore , Indian Music Zonal , Western Music zonal , Drama

    I've been an active participant in extra curricular activity throughout my school life , I've also won Zonal Competition for singing and various interschool competition for singing and extempore

Licenses & Certifications

Volunteer Experience

  • National Service Scheme Graphic

    Volunteer

    National Service Scheme

    - 1 year 1 month

    Environment

    Participated in environmental drive across the campus

  • PayPal Graphic

    Product Owner - Tech for Good

    PayPal

    - Present 3 years

    Economic Empowerment

    Working as a Product owner to empower by teaching/study modules for CDC Lepropsy and Handicapped women on web development, excel and programming languages.

Publications

  • Impact of Binning on Scorecard and Machine Learning models in Credit Risk industry

    BITS PILANI

    Statistical models are key in Credit Risk domain, It is used to assess the risk worthiness of
    the customer which becomes a driving factor in assigning Credit to an individual consumer
    or businesses. Scorecard models are one such model where the underlying model is a Logistic
    Regression where the core objective is to drive explain ability of the model. The transparency in
    model plays a driving factor in Credit Risk field which are driven by several regulatory authorities
    and…

    Statistical models are key in Credit Risk domain, It is used to assess the risk worthiness of
    the customer which becomes a driving factor in assigning Credit to an individual consumer
    or businesses. Scorecard models are one such model where the underlying model is a Logistic
    Regression where the core objective is to drive explain ability of the model. The transparency in
    model plays a driving factor in Credit Risk field which are driven by several regulatory authorities
    and compliance. In recent time, Machine Learning models has picked traction however explain
    ability of the model still remains a big question. The core component of a Scorecard model is
    binning which is discretizing continuous elements into sub-categories, these sub-categories help
    the end-user understand his/her Credit score and also helps in treatments of various strings/nulls.
    The concept of Weight-of-Evidence is used to make these models, However weight of evidence is
    very sensitive to binning.

    See publication

  • Multipath Routing using Machine Learning and Segment Routing and SDN

    SRM University

    The goal of this project was to study segment routing with the help of Machine Learning algorithms and test its traffic engineering capabilities to route the traffic in Multiple paths simultaneously in DC networks by creating a working virtualized testing environment. In addition, another goal was to produce clear instructions for setting up the testing environment and running efficiency tests in it based on different Machine learning algorithms

    Other authors
    See publication

  • Create a model to predict housing prices

    Towards Data Science / Medium

    See publication

  • How to begin your own data science journey

    Towards Data Science / Medium

    Begginers guide to Data science and Regression

    See publication

Patents

  • Interactive System to Build Machine Learning model for Credit Risk model

    Filed 202141036529

    For automated modelling system enabling , rapid Machine learning model development and deployment.

    Patent filled in Indian Patent Office.

    Other inventors

Courses

  • Data Engineering on Google Cloud Platform Specialization

    RUA5TAYUN5EX

  • Data Scientist - Master Degree (Simplilearn)

    288747

Projects

  • Traffic Sign classification using Tensorflow with LeNet Architecture

    Using OpenCV to preprocess the images present in the german traffic sign dataset. The dataset is then trained on a 3 layer Convolution net along with MaxPooling and relu layer. The architecture is based on LeNet Architecture and gives a test accuracy of 95.11%.

    See project

  • Lane Detection for Self Driving Car using OpenCv

    Using OpenCV library in python to detect lanes on the road. The lane detection using technique's such as Canny Edge Detection , Gaussian Blur and Hough transform to process the image. The lanes are detected by extrapolating the lines on the image/video.

    See project

  • Predicting winners of MLB(Major League Baseball) using Lehman2016 Dataset

    Predicting the winners of MLB using ML .

    See project

  • Movie Lens Dataset analysis

    Basic analysis using pandas , matplotlib in python on movielens dataset , such as what gender likes what type of movies , Highest rating and lowest rating.

    See project

  • Using Machine Learning to find optimal/best path for packet transfer through ONOS simulation

    -

    Using Machine Learning on data created by simulation using ONOS to find optimal path for packet transfer depending upon various parameters.

    Other creators

  • Behavioral Cloning

    -

    In this project , We train our computer to drive car using simulator on it's own by processing data generated through driving the car. The data generated is in form of images which are then used as training data. Augmentation is done to increase the training set. VGG and NVIDA end to end models are used and compared in this project.

    See project

  • Recommendation System

    -

    Created recommendation system using python for two different datasets (Movielens and Million Song) , used both collaborative filtering and content-based filtering

Honors & Awards

  • Spot award

    PayPal

    Received Spot award for creation of LGD/CECL and PD models for different portfolios.

  • Spot award

    PayPal

    Was given the Spot award for completion of a critical project which impacts business team for revenue collection. The award was given for a quicker turnaround time and solution!

  • Outstanding Performer Award

    Dun & Bradstreet

    Was given an outstanding performer award for the year 2019-20 , for accomplishing different projects resulting in generation of revenue stream through creation of Analytical products and solutions across Risk and Marketing domain.

Languages

  • English

    Full professional proficiency

  • Hindi

    Full professional proficiency

  • Tamil

    Elementary proficiency

Recommendations received

11 people have recommended Shreyas

Join now to view

More activity by Shreyas

View Shreyas’ full profile

  • See who you know in common
  • Get introduced
  • Contact Shreyas directly
Join to view full profile

Other similar profiles

Explore collaborative articles

We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.

Explore More

Others named Shreyas Raghavan

5 others named Shreyas Raghavan are on LinkedIn

See others named Shreyas Raghavan

Add new skills with these courses

See all courses