Mohammad Ahmad

Mohammad Ahmad

Research Intern at IIITD



I’m a final-year student pursuing B.Tech from Jamia Millia Islamia in Electronic and Communication. Currently doing research internship at IIITD. I have a major interest in programming and problem-solving. I have good knowledge of Machine Learning (And AI) and am interested in Research in ML or DL domains. I have done a good amount of projects (Listed Below) to support my skills. I also have good knowledge and experience in Web development (Both FrontEnd and Backend).


  • NLP
  • Image processing
  • Artificial Intelligence


  • B.Tech in Electronics and Communication, 2021

    Jamia Millia Islamia

  • Senior School Certificate Examination(Class 12), 2016

    Dev Samaj Modern Schoo



Research Intern


Jul 2020 – Present New Delhi
Project: Parameter Estimation in Multi-standard Wideband Receivers via Deep Learning.

SDE Intern


Apr 2020 – Jul 2020 New Delhi
  • Worked on Telemed product with frontend on React.
  • Integrating video calling feature using MS teams API.
  • API and utility development on Django.

Research Intern

WowChemE M&C Pvt. Ltd

Sep 2019 – Jun 2020 New Delhi
  • Explored applications of machine learning in understanding catalysis for energy and chemical production.
  • Successfully reviewed and summarized a few research papers.

Software Developer Intern

Jul 2019 – Oct 2019 New Delhi

Ground transport and logistics solution company based in Germany

  • Worked on Database design and their migrations.
  • Designed multiple API endpoints and their integration.
  • Integrated many services like mailing, logging, stripe, etc.

Recent Publications

MeToo: Sentiment Analysis using Neural Networks

In our work we report our models, their architectures and their respective performances in the Big MM Challenge MeToo. We showed how the ensemble model, which was a collection of 3 different deep learning models, and used two different sets of Embedding vocabularies performed better than the individual models. In the future, we would try to give the problem a more defined approach, given the bias in positive and negative classes for the sentiment classes, we can try to pre-process the dataset in a better way so as to accommodate all training examples, while making sure that the outputs from negative classes do not dominate over the positive outputs from the respective classes. Some other areas which could be improved include, data scraping from the attached urls to the tweets, which can be helpful to find the relevance and judge the authenticity and perform better language understanding.


Regional(India) Winner - Prize worth $2000
Transfer learning model built using RoBERTa deployed on Heroku using Flask and React

  • Sentiment analysis of IMDb movie review using RoBERTa.
  • Used transfer learning from RoBETa pre-trained weights.
  • Transfer Learning helps to improve model with 0.80 to 0.81 f1 score.
  • Deployed on Heroku using React and Flask.
See certificate

TensorFlow: Data and Deployment 4 course Specialization

  • Learned how to save models and use them in web and mobile devices.
  • Learned about TensorFlow splits, tensorboard, and how to share pre-trained models with Tensorflow Hub.
  • And how to create a data pipeline using TensorFlow.
See certificate

Deep Learning Specialization

  • Learned about Deep and shallow Neural Networks.
  • Learned all about activation function, optimization, regularisation, etc.
  • Different strategies like dataset distribution, human-level performance, error analysis, etc.
  • CNN, different architects of CNN like leNet, googLeNet, etc. And some object detection algo like yolo, etc.
  • Learning about RNN, word embedding, sequence to sequence modeling, etc.
See certificate

Programming for Everybody (Getting Started with Python)

See certificate




A Machine Learning Blog, completely built from scratch. It has almost all the features a traditional blog has like commenting, bookmark, likes, auth, etc. Tech Stack - Node.js, React.js, Bootstrap, and PostgreSQL.

Face Recognition Using Tensorflow.js

A Web application to recognize faces and identify it, One can train it by entering the name and their pics in the browser, this is done by using Pre-trained model mobilenet_v1_1 from Tensorflow Hub.
Face Recognition Using Tensorflow.js

Reddit Flair Detection

Detection of popular Reddit flair on “r/india/” subreddit. This project includes data scraping, EDA, modeling, and Heroku deployment.
Reddit Flair Detection

Toxic Text Analysis Using Tensorflow.js

Web Application built on HTML, CSS, and javascript, which analyze toxicity in a given Text using a Pre-trained model from Tensorflow Hub.
Toxic Text Analysis Using Tensorflow.js

Country Detail App

Android Application which show details of all country. Used retrofit library to get json data from given endpoint. Used this API for country data.
Country Detail App







Machine Learning