About
Software Engineer & ML and DL Researcher.
- Birthday: 23rd March
- Phone: +91-94449-78521
- City: Chennai, IND
- Age: 23
- Degree: Bachelor's in Computer Science
- Email: joishbosco99@gmail.com
My name is Joish J Bosco. I work as a software engineer. Speaking about me, I wanted to be a Hacker when I was doing my undergrad. As days went by, I figured out that there was a need to write my tools and, so I began to learn and write my code. Gradually, I got more interested in writing code that could potentially solve problems that I was facing. It was this time I was exposed to Artificial Intelligence. As a hobby and in my free time, I continuously learn and research in Machine Learning, Deep Learning. Something that I would like to add on is that I am a Philomath, always open to challenges and constructive feedback.
Resume
Education
Bachelor's in Computer Science
2014 - 2018
Anna University, Chennai, India
Professional Experience
Junior Software Engineer
2018 - Present
DCKAP Technologies Pvt. Ltd., Chennai, India
- Was a part of the Cloras Back-End Team - Responsible for developing controllers using Flask and MongoDB
- Was a part of the Cloras Front-End Team - Responsible for developing components in ReactJS
- Provided ideas to re-architect the Cloras product.
- Provided ideas to Data Science and Artificial Intelligence team and also worked along with them in building Machine Learning models.
- Was part of the service’s team, assisting in System Integration using Cloras
Research and Development
2018 - 2018
Teknuance Info Solutions Pvt. Ltd., Chennai, India
- Developed multiple Proofs-of-Concepts in order to build Machine Learning capabilities in the team.
- Incorporated various Machine Learning features in Teknuance internal tools, in order to utilize the performance gain offered by Machine Learning algorithms.
- Developed a prototype for AutoML.
- Developed a prototype for converting English text to SQL query.
Skills
Research and Related work
Code Contribution
JsonJ
JsonJ is a simple Python Library that allows parsing Complicated Nested Json tree to a Simple Json data. This Library allows converting Complicated Nested Json Data to an Intermediate Simple Json Data, which make it easier to process the original data.
FeatureSelection
FeatureSelection is a Python Library where you automatically or manually select those features which contribute most to your prediction variable or output in which you are interested in.
- Link
- Currently Working on it
Published Work
User Independent Human Stress Detection
2020 IEEE 10th International Conference on Intelligent Systems (IS)
Ramanathan Murugappan, Joish J Bosco, Krishnakanth Eswaran, Pranav Vijay, Vineeth Vijayaraghavan
Stock Market Prediction and Efficiency Analysis using Recurrent Neural Network
GRIN Verlag
Joish Bosco, Fateh Khan
Currently Working Research
A Two-Stage Machine Learning Architecture for Malware Detection and Data Theft Classification
IEEE Consumer Communications and Networking Conference (CCNC)
Joish J Bosco, Prasanna Kumaran D, Sai Vishwanath Venkatesh, Pravin Kumaar R, Vineeth Vijayaraghavan
- Awaiting Author Notification
Detection of Malicious activity based on mobile phone usage over concept drift
- Currently Working on it
Personal Projects
Malware Detection and Data Theft Classification
- In this project, We as a team were able to predict different malware attacks and the different data that was stolen during an attack.
- We used Total memory used size, Number of threads that were running, and some more features that were a part of the Sherlock dataset.
- NOT YET PUSHED TO GITHUB - (Will be done soon)
Human Stress State Detection
- In this project, We as a team were able to figure out if a person was in a stress state or not.
- We used Electrocardiogram (ECG), Electroencephalogram (EEG) signals that were a part of the Wearable Stress and Affect Detection (WESAD) dataset.
- Link
Stock Market Predictions
- In this project, I was using Long Short Term Memory (LSTM) to forecast the future stock price of Google.
- Link
Automated Mailing System
- In this project, I was able to figure out the absent list of students and sent a mail to there parents/guardians informing them.
- This complete process was automated.
- NOT YET PUSHED TO GITHUB - (Will be done soon)