• Hi!
    I'm Atharva

  • I am
    CS PhD @ ASU

About Me

Who Am I?

Hi I'm Atharva Khedkar, I am a PhD in Computer Science student at Arizona State University, where I conduct cutting-edge research on applying Artificial Intelligence to make Machine learning applications faster using accelerators and MLIR. This aligns with my passion for building artificial intelligence software that can enhance human-machine interaction and create better experiences for various industries.

Before joining ASU, I worked as a Machine Learning Intern at AMD and a SDE Intern at HackerRank, where I gained valuable experience and skills in developing and deploying scalable and robust MLOps pipelines using Kubernetes and Docker for large-scale NLP projects, and implementing and testing several machine learning and deep learning algorithms using Python, TensorFlow, and Microsoft Azure. I also have a track record of winning multiple national and international awards and honors, such as being a 2x Microsoft Imagine Cup - National Champion & World Finalist, a Gold Microsoft Learn Student Ambassador, and a Winner of the Accenture AI Hackathon. I have a strong background in artificial intelligence, machine learning, and deep learning, with proficiency in Python, TensorFlow, Microsoft Azure, Kubernetes, and MLOps, and familiarity with SQL, R, Computer Vision, and NLP. I enjoy learning new things and connecting with people across a range of industries, so don't hesitate to reach out if you'd like to get in touch.

Compiler Development

Deep
Learning

ML Accelerator Design

Artificial
Intelligence

My Specialty

My Skills

MLIR

97%

Deep Learning

95%

Machine Learning

92%

Microsoft Azure

80%

Signal Processing

75%

TensorFlow

90%

LLVM Project

89%

Flutter

85%

Backend Development

90%

Ruby on Rails

88%
Education

Education

  • Arizona State University
  • Graduation: June 2028 (Expected)
  • CGPA: 3.75/4
  • Developing a Digital Signal Processing Dialect using Multi-level Intermediate Representation (MLIR).
  • G H Raisoni College of Engineering, Nagpur
  • Graduation: June 2023
  • CGPA: 9.25/10
  • Branch Topper/ Gold Medalist - Freshman and Junior Year
  • Teaching Assistant - Fall 2021, for the course Artificial Intelligence Knowledge Representation and Reasoning
  • G H Raisoni Vidyaniketan, Nagpur
  • Senior Secondary | CBSE
  • Graduation: 2019
  • Qualified in top 10 out of 1300 students in Nagpur's Best Student Contest 2017 organized by Rao IIT Academy
  • Vimaltai Tidke Convent, Beltarodi
  • CGPA - 10/10 | CBSE
  • Graduation: 2017
  • Ranked in top 250 out of 3000 students in All India Personality Contest 2016
Experience

Work Experience

Graduate Research Assistant, ASU Aug 2023 - Present

  • Developing a Digital Signal Processing Dialect using Multi-level Intermediate Representation (MLIR).
  • Implemented canonical optimizations and lowerings of DSP applications into Affine and LLVM dialects.

ML Intern, AMD Feb 2023 - July 2023

  • Worked in Integration & Validation (IV) team and created ML networks to automate graphics driver fault detection.
  • Successfully reduced test planning time by 80%, resulting in faster turnaround for driver releases while maintaining high-quality standards.

SDE Intern, HackerRank Jul 2022 - Dec 2022

  • Analyzed various ML Pipeline orchestration tools such as Pachyderm and MLflow and deployed AWS Lambda function to automate model deployments from MLflow Model registry into Sagemaker endpoints.
  • Developed and Launched Application Tracker feature used by 1000+ developers on the HackerResume project.

Research Intern, University of Cambridge Jun 2021 - Oct 2021

  • Principal Investigator: Dr. Hatice Gunes
  • Worked on Deep Transfer Learning Techniques for Affect Analysis in Work-like Settings at Affective Intelligence and Robotics Lab.
  • Developed deep learning models for the WorkingAge project to investigate spatial and temporal changes in facial affect (valence and arousal) in work-like situations.

Consultant, Cybercrime Cell Nagpur Jan 2021 - Apr 2021

  • Provided consultancy to private companies and government agencies in the fields of data analysis, artificial intelligence, machine learning and cloud computing.
  • Conducted data analysis using Tableau and Microsoft Excel on the cyber-crime cases registered in the city in the year 2020 for Cyber Crime Cell, Nagpur City Police.

Chairperson, Entrepreneurship Cell - GHRCE Sep 2021 - Aug 2022

  • Promoted Startup culture while focusing on Female Entrepreneurs in Technology by organizing B-plan and business development workshops.
  • Assisted in the incubation of 7+ startups and 42+ students at the Campus Business and Incubation Center (GHRTBIF).

Vice-Chair, IEEE CIS SBC GHRCE Jan 2021 - Dec 2021

  • Organized and Mentored 10+ events with 3000+ participants from all over the world.
  • Under my vision, IEEE CIS SBC received the " 2022 Outstanding Chapter Award" by IEEE CIS Awards Committee with the citation “For the innovative and wide range of diverse and inclusive activities which connect research, education and industry in the field of computational intelligence.”

Microsoft Learn Student Ambassador April 2020 - Current

  • Gold Student Ambassador
  • Trained more than 500 students by conducting a machine learning bootcamp as a student ambassador.
  • Conducted 20+ sessions on open source, GitHub and mentored 500+ students in MLH Hackathons.

My Work

Projects

Detect-d

Detect-d is a deepfake detection platform that uses the same technology to detect deepfakes that is used to create them- Artificial Intelligence.

AutoML - Automatic Machine Learning App

Created a Automatic Machine Learning App which takes a .csv file as an input and returns best machine learning model comparison results to the user within minutes.
Deployed with Azure App services and Integrated Azure Pipelines CI/CD with GitHub.

ScaNet

ScaNet is a web platform which classifies between 14 different lung diseases using Chest Radiographs.
I created a multi-label classification model using transfer learning and trained it on the NIH Chest X-ray dataset and Pneumonia Chest x-ray dataset resulting with 94.2% validation accuracy. ScaNet is deployed on Microsoft Azure and Firebase and includes features like finding doctors nearby and training custom models using intuitive web interface.

MedScan

MedScan is a modern smartphone and web based electronic health record system that helps users to save their medical reports, prescriptions digitally at one place for the healthcare industry here in India.The solution is a Online-to-Offline (O2O) solution and includes use of modern technologies for operations.

Spamia- SMS Spam Clasifier

Created a SMS Spam Classifier using Mutinomial Naive Bayes Algorithm.
Deployed using Flask and Heroku.

Get in Touch

Contact

social social social social social