• Howdy!
    I'm Avinash

    Machine Learning Engineer
    Java Developer

    Download Resume

About

Who Am I?

Hey, I'm Avinash Shanker a computer science graduate student at University of Texas, Arlington graduating summer 2020. I am interested by complex problems in Artificial Intelligence, specifically those which require optimization and machine learning. I worked as a Software Engineer at TESCO for span of 3 years before deciding to pursue my masters. At TESCO, I was part of team responsible for applications which dealt with creating real-time ordering, stock replenishment and allocation of products to stores and warehouses in Thailand, Malaysia and Europe. Graduated with B.Tech in Information Technology at National Institue of Technology, Karnataka (ranked 8th among India's engineering schools). I enjoy generating new ideas and devising feasible solutions to broadly relevant problems. My colleagues would describe me as a driven, resourceful individual who maintains a positive proactive attitude when faced with adversity. Currently, I am seeking opportunities that allow me to develop Java APIs for Machine Learning models.

I aim to make a difference through my creative solutions.

What I do?

Here's area my expertise

Machine Learning

Deep Learning

Java Developer

Education

Alma Mater

University of Texas at Arlington, Texas

Dual Specialization: Artificial Intelligence and Database Systems

  • Duration : 2 years (2018 - 2020)
  • GPA : 3.67/4.0
  • Courses :
    • Neural Networks
    • Computer Vision
    • Artificial Intelligence
    • Big Data(Cloud)
    • Data Mining
    • Advanced Algorithms
    • Parallel Processing
    • Distributed Systems
    • Software Testing
National Institute of Technology Karnataka, Surathkal
  • Duration : 4 years (2011 - 2015)
  • GPA : 3.5/4.0
  • Courses :
    • Automata and Compiler Design
    • Mobile Computing
    • Computer Graphics
    • Operating Systems
    • Semantic Web Technologies
    • Human Computer Interaction
    • Information Assurance & Security
    • Cloud Computing
    • Data Warehousing and Data Mining
    • Wireless Sensor Networks
    • Unix Programming
    • Microprosessors and Interfacing
Kendriya Vidyalaya Hebbal, India

All India Senior School Certificate Examination(XI to XII)

  • Duration : 2 years (2009 - 2011)
  • Percentage : 90.00%
Experience

Professional Work and Internship

Software Engineer  Jul 2016 - Jul 2018

TESCO Bengaluru

  • Developed RESTful web service using Spring framework and MongoDB to monitor and re-process orders which failed to integrate in Store system. Used ReactJS UI to consume the service and provide feedback on order rejection status. This enabled to easily correct orders, reducing manual effort on integrating to store from 6hrs/week to 0.5hrs/week
  • Designed a multithreaded mechanism to enhance the performance of existing auto-ranging functionality in Product Information Management (PIM) system. Auto-ranging assists in ranging 50million products to newly setup warehouse. To expedite, distributed load evenly to 10-threads by hashing product ID. Reduced job runtime of 6hrs by 80%
  • Developed an API using Spring framework to validate, daily sales with current stock on hand. Any discrepancy in expected stock, enabled Stock Check API to flag such products in the DB for manual review and generating notifications on dashboard. This enhanced the response time for stock discrepancy correction by 90%

Graduate Software Engineer  Jul 2015 - Jun 2016

TESCO Bengaluru

  • Developed React native Android and iOS application for TESCO shopper's in U.K. which pushes location based customized promotions and deals to users. App sends notifications based on previous purchases when in vicinity of a Brick and Mortar TESCO store
  • Was part of team responsible for handling Product metadata ranging over a million and maintaining stock ledger of these products
  • Proficient in functional knowledge and workflows of Retail applications
  • Worked with API monitoring, alerting & version control tools including Git, Splunk, Jenkins and App dynamics
  • Expertise in working Java, ProC, Triggers in PL-SQL and Shell Scripting
  • Worked with RMS, RIB and Integration applications to ensure steady realtime flow product data across systems for Tesco International Clothing Brand (TICB), Brick & Mortar stores in Thailand, Malaysia and Central Europe(CZ, SK, PL & HU)

Engineer Intern   May 2014 - Jul 2014

Bharat Electronics Limited

  • Was part in developing prototype front end graphical display termed Radar Soft Scan Simulator. It was used for advanced detection and display of clutter objects within a specific target signature
  • Implemented functionality using openCV libraries in python

Achievement

Awards

TESCO Value Award for effectively coordination with Business Team in Europe for auto-ranging over half a million foundation data on products while setting up of multiple new stores in Czech and Poland. This was a critical as Items cannot be ordered or sold otherwise, adding direct financial value to business

  • Honored On : Jan 2018

TESCO Star Performer Award for innovation and dedicated work on BOL, allocation and Transfer Reject reprocessing among team of 20, this automation helped in reducing 14-man hrs/week taken by business team to do it

  • Honored On : Mar 2017
Certification

Certification

Java EE8: Web Services  Mar 2020

Linkedin Learning

Develop modern and lightweight web services using Java Enterprise Edition (EE) 8. Instructor Tayo Koleosho begins by providing some context, explaining why we develop web services, how SOAP and REST differ, and what's new in Java EE 8. He then shows Java developers how to implement SOAP web services and RESTful APIs, such as how to design and develop a RESTful service using the Java API for RESTful Web Services (JAX-RS). Learn about testing with Postman, validation and error handling in JAX-RS, logging and monitoring in JAX-RS and JAX-WS, API security using JSON Web Token (JWT), and more.

Certificate & Credentials

Designing RESTful APIs Certification  Feb 2020

Linkedin Learning

Having a solid understanding of how to correctly build APIs and creating websites. Learn how to plan and model your own APIs, and explore the six REST design constraints that help guide your architecture. Identifying the users or "participants" of system, and the activities they might perform with it. Validating design before we build it, and explore the HTTP concepts and REST constraints needed to build API.

Certificate & Credentials

Advanced Java Programming  Feb 2020

Linkedin Learning

Course covered variety of topics, including generics, working with the Collections Framework, and functional programming. Plus, learn about I/O in Java, working with files and directories, and structuring applications using the modular system available in Java.

Certificate & Credentials

Work

Academic And Leisure Projects

Deep Privacy Face De-Identification Using Generative adversarial networks

Developed a model which can change a persons face to look like a completely different person, thus protecting their privacy. The model is trained with combination of generative adversarial networks(GAN) and autoencoders. Model ensures anonymity by synthesizing GAN generated images. The generated faces are used to de-identify subjects in images or video.

LINK

Google Inception Convolution Neural Network For Image Classification

Designed & trained a 3 layer densely connected CNN which was visualized using TensorBoard. Used deep learning network for image classification generated an accuracy of 86.4% on ideal hyperparameters. cPerformed on CIFAR10 dataset which consists of 60,000 images using TensorFlow, Keras, cv2, pandas, NumPy lib

LINK

Read

Recent Projects

Nov 23, 2019 | Deep Learning

DEEP PRIVACY FACE DE-IDENTIFICATION USING GENERATIVE ADVERSARIAL NETWORKS

Developed a model which can change a persons face to look like a completely different person, thus protecting their privacy. The model is trained with combination of generative adversarial networks and autoencoders. Model ensures anonymity by synthesizing GAN generated images. The generated faces are used to de-identify subjects in images or video.

Oct 31, 2019 | Deep Learning

GOOGLE INCEPTION CONVOLUTION NEURAL NETWORK FOR IMAGE CLASSIFICATION

Designed & trained a 3 layer densely connected CNN which was visualized using TensorBoard. Deep learning network for image classification generated an accuracy of 86.4% on ideal hyperparameters. Performed on CIFAR10 dataset which consists of 60,000 images using TensorFlow, Keras, cv2, pandas, NumPy lib

Oct 17, 2019 | Deep Learning

Data Analysis On Drug Consumption Dataset Using K-Means & Agglomerative Clustering

Determined number of clusters required for dataset and reduced essentials using principal component analysis. Performed K-Means and Hierarchical Agglomerative clustering to train, predict and contrast the performance of both models using confusion matrix (precision & recall) then visualizing the data with 92.3% accuracy in K-means and 79% in agglomerative clustering. Implemented using pandas, scikit, NumPy libraries in python

Get in Touch

Contact

avinash.shanker@mavs.uta.edu

Dallas, Texas

+1-682-401-9365

Follow me on :