About
I am a Software Engineer specializing in Backend Development, Deep Learning and Machine Learning.
I have a keen interest in software development and love to work on different applications. I am a driven and passionate individual,
interested in working on intriguing ideas with enthusiastic team in a competitive and collaborative environment
Looking for Software Engineering and Machine Learning Engineer roles
Skills
Resume
Professional Experience
Salesforce - Associate Memeber of Technical Staff (July 2021 - Current)
Working with the Revenue Cloud team.
Samsung R&D Institute, Noida - Software Engineering Intern (June 2020 - November 2020)
Working under the Project titled "Handwriting Recognition from Medical Precriptions" with the R&D team at Samsung Noida and Samsung Digital Academy Research Lab DTU.
Delhi Technological University - Undergraduate Researcher (August 2019 - June 2020)
Working under the Project titled “Software Bug Predictor” by Principal Investigator, Dr. Ruchika Malhotra, Professor, and Associate Head, Department of Computer Science and Engineering, Delhi Technological University.
The Energy and Resources Institute India - Software Engineering Intern (December 2019 - January 2020)
As a Software Engineering Intern, involved in data science, analysis and Multi variate Time Series prediction of the Air Quality Index with LSTM's and Recurrent Neural Networks in Python. Used Tensorflow, Keras and python libraries like pandas, numpy, sci-kit learn, seaborn etc for data preprocessing and manipulation.
Education
Bachelor of Technology - Software Engineering (2017 - 2021)
Delhi Technological University, Delhi, IN
GPA : 8.9
AISSCE - PCM Computer Science (2005 - 2017)
Lancer's Convent Sr. Sec. School, Delhi, IN
Aggregate : 92%
Research Projects
Handwriting Recognition for Medical Prescriptions using a CNN-Bi-LSTM Model
Devised a medical prescription detection technique based on Deep Learning. Built a CRNN model with Connectionist Temporal Classification. We also devised a data preprocessing technique, wherein the input image is grayscaled, blurred(gaussian) and dilated to extract text line by line, which is fed as inputs to the deep learning model. A full paper has been accepted at the IEEE 6th International Conference for Convergence in Technology
Software Bug Predictor
Built a Software Bug Predictor analysing Java based projects, collecting data from GitHub repositories, classifying information as Java CKJM metrics. Tested on over 20k commits. Conducted research and development and wrote scripts in Python for data collection, manipulation, and machine learning models used in prediction, reducing development time by up to 20%. Built a front-end using Electron on Linux/Unix environment.
Frameworks
Python
C++
Machine Learning
Deep Learning
Tensorflow
Azure DevOps
Android Studio
HTML5
CSS3
PostgreSQL
Git
Google Cloud Platform
Portfolio
- All
- Deep Learning
- Research Experience
- Android Development
- Backend Development
Handwriting Recognition using Deep Learning
Devised a medical prescription detection technique based on Deep Learning. Built a CRNN model with Connectionist Temporal Classification. We also devised a data preprocessing technique, wherein the input image is grayscaled, blurred(gaussian) and dilated to extract text line by line, which is fed as inputs to the deep learning model. A full paper has been accepted at the IEEE 6th International Conference for Convergence in Technology
Part of Speech Tagging - NLP
Used the Pomegranate library to build a hidden Markov model for part of speech tagging, with an accuracy of 97%. Part-of-speech tagging, also called grammatical tagging, is the process of marking up a word in a corpus as corresponding to a particular part of speech, based on both its definition and its context.
Automatic Speech Recognition
Built a deep neural network that functions as part of an end-to-end automatic speech recognition (ASR) pipeline. The model converts raw audio into feature representations, which will then turn them into transcribed text.
Software Bug Predictor
Built a Software Bug Predictor analysing Java based projects, collecting data from GitHub repositories, classifying information as Java CKJM metrics. Tested on over 20k commits. Conducted research and development and wrote scripts in Python for data collection, manipulation, and machine learning models used in prediction, reducing development time by up to 20%. Built a front-end using Electron on Linux/Unix environment
Image Caption Generator
Developed an image caption generator that generates captions for the provided image. Used CNN’s as encoders to extract features from the image and RNN’s as decoders to do language modeling.
Trigger Word Detection
Built a Trigger Word Detection Model using convolutional Layers and Gated Recurrent Units. It is based on Natural Language Processing, the technology that allows devices like Amazon Alexa, Google Home, and Apple Siri to wake up upon hearing a certain word. Building an Android application for the same.
Take Care
I am developing an Android application that detects user’s mental health based on their search and word choices. The application would analyse the user word choices with the help of on-device Machine Learning to calculate the results. A record of what user’s search relates to (being depressed or not depressed) will be saved on the device, and an active graphical analysis could be viewed for it.