About Abhishek
Abhishek is currently working as a Data Scientist at Autodesk. He has a Masters in Information Science (MSIS) from the University of Pittsburgh, Pennsylvania, USA. At Autodesk, he's working with the Digital Experience and Customer Empowerment (DXC) Data Science Team in improving the user's search experience in Autodesk's ECommerce platforms. He has previously worked as a Machine Learning Intern at the UPMC Hillman Cancer Research Center (MoSHI Lab), applying complex machine learning algorithms to accelerate oncology research and improve the lives of patients suffering from cancer. Abhishek has earned his bachelors degreee in Electronics and Communication Engineering from the Vellore Institute of Technology (VIT), Chennai Campus, India. He has a two year experience as an undergraduate researcher working on deep neural networks for application in facial emotion recognition and speech recognition. Being passionate about machine learning and data analytics, he has deeply involved himself in applying his theoretical knowledge in developing and analyzing solutions for real world problems affecting humanity.
He always had a profound passion and fascination for areas requiring analytical approach and critical thinking. Right from early days at school, mathematics and science subjects have intrigued him. Visualizing things and quick learning helps him keep updated with the current trends in artificial intelligence domain. He is a highly motivated and meticulous individual who loves to develop and train accurate and efficacious machine learning models.
Programming Skills
-
C
90% -
C++
90% -
Python
80% -
Data Structures
80% -
Java
60%
Frameworks
-
TensorFlow
70% -
Scikit-Learn
70% -
Spacy
60% -
NLTK
60% -
Haystack
40% -
Flask
40%
Tools
Jupyter
Pandas
Spyder
Hugging Face
Heroku
AWS EC2
AWS Sagemaker
Haystack
Google Cloud
MongoDB Atlas
SQL
AIR-FLOW
Apache Spark
PostgreSQL
GitHub
Postman
NetSim
Proteus
VS Code
Soft Skills
-
Hindi
100% -
English
90% -
French
40%
Resume
Working History
-
Data Scientist
Jan 2023 - Present
-
Data Science Intern ( eCommerce )
Jun 2022 - Sep 2022
• Working on improving Autodesk’s search bar by implementing semantic based Dense Passage Retrieval with Generative Question Answering functionality over traditional TF-IDF retrieval methods. Implementing Haystack framework to organize a robust search system pipelines.
• Generated Facebook Ai Similarity Search (FAISS) document store on Amazon S3 by pre-processing 3.5 million data points to generate and store vector embeddings for efficient similarity searches. Developing Streamlit based front end for interactive query tuning.
-
Data Science Intern ( Machine Learning )
Jul 2021 - May 2022
• Developed predictive machine learning models to predict remission in patients going through cancer treatment based on activities of daily living (ADLs).
• Worked on AWARE mobile passive sensing framework to gather patient’s daily life activity data and digital bio-markers. Used RAPIDs data pipelines to stream mobile and wearable device data to extract behavioral features for analysis and training predictive models.
-
Research Data Analyst ( Biomedical Informatics )
Mar 2021 - Present
• Working as a Research Data Analyst, helping develop reports, analysis and visualizations. Managing and designing the reporting environment, including data sources, security, and metadata. Providing technical expertise in data storage structures, data mining, and data cleansing.
•Collaborative Institutional Training Initiative (CITI Program) certified researcher for Biomedical, Privacy and Information Security, Responsible Conduct of Research, Social-Behavioral Educational Course and Conflict of Interest
-
Undergraduate Researcher (Deep Learning)
Jul 2018 - Apr 2020
• Worked on Deep Learning with emphasis on developing new hidden layer architecture and algorithms for facial emotion recognition in Convolutional Neural Network (CNN).
• Implemented Convolutional Neural Network (CNN) for Speech Recognition System using Mel-frequency cepstral coefficients (MFCC).
-
IEEE Student Chapter Technical Head
Apr 2018 - Nov 2019
• Worked as the technical head for student chapter for the past one year, looking after all the technical activities for organising Hack-a-Thons, Code-a-Thons & Project Expos.
•Assisted chapter members towards new innovative project development and research works in Signal Processing and Machine learning.
-
Foreign Technical Training Program
May 2018 - Jul 2018
• Developed intelligent data routing algorithms for the Vehicular Ad. Hoc. Networks (VANET) in autonomous cars at the advanced technological center (ATC)
• Worked along with ’VicRoads’ in developing smart solutions to reduce traffic congestion in Melbourne Metropolitan Area.
Education History
-
Masters in Information Science (MSIS)
Jan 2021 - Dec 2022
GPA: 3.9/4
Pursuing Masters in Information Science with concentration in Big Data Analytics.
Courses: Machine Learning, Data Mining, Data Centric Computing, Artificial Intelligence, Database Management Systems, Algorithm Design & Analysis -
Bachelors of Technology (B. Tech.) in Electronics and Communication Engineering
Jul 2016 - May 2020
CGPA: 8.95 / 10
Among the top 10% of students in the batch.
Courses: Neural Networks, Computer Networks, Data structures and Algorithms, Embedded C and Linux, Signal and Image Processing, Embedded System Design.
Extra-Curricular
Organiser & Tech. Head - CodeBuzz
TECHNO-VIT 2018
Organised online coding contest during the technical fest Techno-VIT 2018 with participation from all over the country.
Organiser @ General Tech. Quiz
IEEE DAY 2018
Organised the Inter Branch quiz contest during the IEEE day 2018 and helped student show their quizzing talent over a broad range of topics.
Member @ Code-Y-Gen
Game Development Club
2017-18
The club helped students learn about game development in UNITY game engine and teach students to develop in and outs of competitive coding skills.
Research Publications
- Abhishek Verma, Piyush Singh, John Sahaya Rani Alex. Modified Convolutional Neural Network Architecture Analysis For Facial Emotion Recognition. In International Conference on Systems, Signals and Image Processing (IWSSIP), Osijek, Croatia. IEEE, 2019. 'DOI: 10.1109/IWSSIP.2019.8787215'
- Md Amaan Haque, Abhishek Verma, John Sahaya Rani Alex, Nithya Venkatesh. Experimental Evaluation of CNN Architecture for Speech Recognition. In International Conference on Sustainable Technologies for Computational Intelligence (ICTSCI), Jaipur, India. Springer, 2019. 'DOI: 10.1007/978-981-15-0029-9'
- Abhishek Verma, Piyush Singh, John Sahaya Rani Alex. Disease and pest infection detection in coconut tree through deep learning techniques. Computers and Electronics in Agriculture Journal, SPRINGER 2020. 'https://doi.org/10.1016/j.compag.2021.105986'
- Sara Chokshi, Yalini Senathirajah, Vandana Yadav, Mimi Winsberg, Erin O’Callaghan, Scott Sullivan, Abhishek Verma, A Comparative Evaluation of Measurement-Based Psychiatric Care Delivered via Specialized Telemental Health Platform Versus Treatment As Usual: A Retrospective Analysis. Cureus 2022. 'https://doi:10.7759/cureus.21219'
Portfolio
Disease Detection in Rice Paddy
Oct 2019 – Dec 2019
Neural Networks
Used the Convolutional Neural Network (CNN) & Deep Neural Network (DNN) to compile and develop a machine learning model that detects the disease in rice paddy based on an input image of the crop with initial symptoms.
Modern-Art Generation via Generative Adversarial Networks (GANs)
Sep 2019 – Dec 2019
Neural Networks
Used Generative Adversarial Networks (GANs) to build a deep-net that was capable of not only learning a distribution of the style and content components of many different pieces of art, but was also able to novelly combine these components to create new pieces of art.
Score Prediction System for IPL (Cricket League)
Aug 2020 - Sep 2020
Machine Learning
Sports Analytics
Score prediction System for Indian Premier League (IPL - Cricket tournament) using last 10 years of ball by ball dataset. The regression model works on 50+ features including stadium track record, past performances of teams against each other etc. Pragmatic RMSE value was achieved. The web application was deployed on Heroku cloud server using Python Flask framework.
Student Stock Trading Platform ( PiTT NiVESH )
Jan 2021 - May 2021
Other
Pitt Nivesh is web app/site that allows the user to connect with friends that might want to combine funds with them, and to use that bigger pot of money to purchase stocks and watch it grow in value. Flask web framework was used to deploy the portal along with daily stock information streaming from the financial stock APIs and MySQL was used as the backend database to store user based information
Interactive Chat Bot
Jan 2020 - Jun 2016
Deep Learning
The chatbot was trained on the dataset which contained categories (intents), pattern and responses. Special recurrent neural network (LSTM) was used to classify which category the user’s message belongs to and then gave a random response from the list of respons.
Cloud Based Paid Car Parking System
Sep 2016 – May 2019
Embedded Systems
Designed a car parking system for university parking using RFID tags, IOT along with Cloud based database managment system to automatically generate invoice monthly.
Voice Recognition Systems using MATLAB
Dec 2017 – May 2018
Signal Processing
Developed a voice recognition system using MFCC and DTW for higher accuracy and improved results in MATLAB. The project was further developed to control a RF car using voice commands.
Augmented Reality (AR) shopping mall App
Jan 2018 – Mar 2018
Augmented Reality
App. Development
Designed & developed an Augmented Reality app for shopping malls using Unity Engine and Vuforia development package. The app lets you scan any product and displays product info in an augmented reality environment and lets you update your shopping cart.
Solar Tracking System
Sep 2015 – Jan 2016
Semiconductor Devices
Developed a solar tracking system to increase the production efficiency of solar energy by more than 20% in traditional solar panels.The solar panels were designed along with LDR sensors to track the position of sun and rotate the panel perpendicular to the sun rays.
Contact Informations
- E-mail: svermaan@gmail.com
- Skype: svermaan