Summary
Overview
Work History
Education
Skills
Certification
Timeline
Projects
Research Paper
Hi, I’m

Abhishek Mohapatra

Machine Learning Engineer
Bengaluru,Karnataka
Abhishek Mohapatra

Summary

Enthusiastic Machine learning and Deep Learning practitioner eager to contribute to team success through hard work, attention to detail and excellent organizational skills. Clear understanding of Machine learning and Deep learning algorithms and worked on various projects. Motivated to learn, grow and excel in the Artificial Intelligence industry.

Overview

4
Certificates
4
years of post-secondary education
6
years of professional experience

Work History

Scaledge (Client: SONY)
Bengaluru, Karnataka

Machine Learning Engineer
12.2021 - Current

Job overview

  • Worked on various modules of the perception based AI engine like visual emotion detector, speech emotion recognizer and speech enhancer.
  • Achieved a validation accuracy of 84.6 % and an inference time of 14.2 seconds and the model was made deployable by optimizing it through Knowledge Distillation or the visual emotion detector and was deployed with Tensorflow JS.
  • Achieved a validation accuracy of 74.2 % and an inference time of 11.6 seconds and the model was made deployable by optimizing it through model pruning for the speech emotion recognizer and was deployed with StreamLIT.
  • The speech enhancer was built using an UNET and TFGAN based network called VoiceFixer which was trained using a custom inhouse dataset and the prototype webapp was made by deploying the model with Gradio .

White Rabbit Designers
Chennai, Tamil Nadu

Data Analyst
01.2020 - 12.2021

Job overview

  • Created and improved intelligence resources to facilitate consistent data management and version control strategies.
  • Composed production-grade code to convert machine learning models into services and pipelines to be consumed at web-scale using FAST API and FLASK.
  • Incorporated Time Series data to understand and translate complex business problems into machine learning problems.
  • Identified and documented project constraints, assumptions, business impacts, risks and scope exclusions which helped in reducing investment by more than 20%
  • Prototyped machine learning applications and quickly determined application viability using web frameworks like Gradio and cloud frameworks like Heroku.
  • Created customized applications to make critical predictions, automate reasoning and decisions and calculate optimization algorithms.
  • Automated ML models using Pycaret, MLFlow and DVC
  • Used Hydra to manage YAML configuration files and automated code review process with Black, Flake8, Isort and Iterrogate in the Pre-commit and Git-Integration .
  • Productionized the ML applications in Azure Container with Docker

Freelancer ( Level App Innovations, Manya Education, Manhattan Review)
Bengaluru, Karnataka

Statistics and Machine Learning Instructor
01.2017 - 09.2019

Job overview

  • Taught and assisted students in areas such as Statistics, Calculus, and ML algorithms, creating tailored lesson plans and study guides on subject matter which helped me improve my own critical reasoning which was important for an aspiring machine learning engineer.
  • Had a huge improvement in my mathematical intuition while teaching these aspirants and got an opportunity to work on various ML frameworks and explore across multiple domains

Education

National Institute of Technology
Rourkela, Odisha

Bachelor of Technology from Computer Science And Engineering
08.2012 - 07.2016

Skills

Machine learning

undefined

Certification

Data Science Math skills Certification, Duke University

Timeline

Machine Learning Engineer

Scaledge (Client: SONY)
12.2021 - Current

Computer Vision Certification, Udemy

06-2021

Machine Learning Certification, Udemy

08-2020

Deep Learning Certification, Udemy

05-2020

Data Analyst

White Rabbit Designers
01.2020 - 12.2021

Data Science Math skills Certification, Duke University

03-2018

Statistics and Machine Learning Instructor

Freelancer ( Level App Innovations, Manya Education, Manhattan Review)
01.2017 - 09.2019

National Institute of Technology

Bachelor of Technology from Computer Science And Engineering
08.2012 - 07.2016

Projects

1. Reinforcement Learning Approach to Self Driving Cars:
In this project our aim is to automate the self-driving car to make round trips between the left top end and right bottom. by training a deep q-learning agent.

https://github.com/Abhi-899/Reinforcement-Learning-Approach-to-Self-Driving-Cars-Deep-Q-learning

2. Real Time Object Detection using YOLOv4:
We train the YOLOV4 network on 3 classes 'Ambulance' , 'Car' , 'Person' with the Google open image dataset and run the detection on a real time video caught on a moving traffic camera.

https://github.com/Abhi-899/YOLOV4-Custom-Object-Detection

3. Lane Detection in a controlled environment using pixel summation and thresholding :
This project is one of the important modules of the complete autonomous cars project where we will try to visualize the lanes that are needed to be traversed along with the amount of curvature at each turning.

https://github.com/Abhi-899/Lane-Detection

4. Behavioral Cloning for Autonomous cars using the DAVE-2 network created by NVIDIA :
Our goal is to teach our Car how to make a lap, using a part of the 1st training track included in the simulation environment given by Udacity. We want our neural network to drive a bit straight, and then make some turns to the right until it reaches the initial point.

https://github.com/Abhi-899/Autonomous-Cars


Research Paper

Publication- Automation of stock forecasting using technical indicators, ISSN:0378-4568

Implemented and compared various Machine Learning, Deep Learning, Reinforcement Learning models used for Time series forecasting and tried to improve efficiency by integrating technical Indicators

Similar Profiles

CREATE PROFILE
Abhishek MohapatraMachine Learning Engineer