Suraj Gattani

Cambridge, MA ยท surajgattani.g@gmail.com

I am a machine learning engineer passionate about cutting-edge technology and solving real-world problems. I am seeking an opportunity with a high growth-oriented organization to put in my best endeavors to translate innovative ideas into feasible projects by applying my technical skills improving the current scenario.

Experience

Machine Learning Engineer - GV20 Therapeutics

  • Deep Learning model architecture implementation for cancer type classification (MCC: 80%) and target gene expression (PCC: 0.50).
  • Applied Transfer Learning and Multitask Learning to improve the expressed target gene correlation by up to 40%.
  • Used AWS Sagemaker to train and optimize the model parameters with the hyperparameter optimization search.
  • Created a python package inclusive of various models from recent antibody-related research.

    • Research Engineer - Bioinformatics and Machine Learning Lab

  • Data preprocessing to analyze the risks and mishaps and build a forecasting model with 97% accuracy.
  • Developed a tagging based system (94%) to forecast the risks according to the location and type.
  • Created a software plugin to automate the risk forecasting and model update with new data.

    • Research Engineer - University of New Orleans

  • Developed novel algorithms for different single/multi-class classification problems.
  • Skilled at detecting, resolving issues, and understanding and analyzing data.
  • Published and/or presented research ndings at national conferences.

    • Projects

    NASA Risk Analysis using NLP

    Implementing deep learning models like CNN, RNN, LSTM, and Spacy for NLP by mining large text data. Creating a software plugin for the test, report, and further investigations.

    Role: Deep Learning Engineer

    NASA Patent and Software Classier tool

    The objective of this project is to systemize NASA's patent and software management, by a set of automated machine learning tool. The Patent and software contain 15 categories each which further need to be divided into sub-categories for efficient classification. The useful features have been extracted from the documents and data provided by NASA as well as from some other data sources.

    Role: Machine Learning Engineer

    StackCBPred - A Stacking based Prediction of Protein-Carbohydrate Binding Sites from Sequence

    A balanced predictor, StackCBPred, which utilizes features, extracted from evolution-driven sequence profile, called the position-specific scoring matrix (PSSM) and several predicted structural properties of amino acids to effectively train a Stacking-based machine learning method for the accurate prediction of protein-carbohydrate binding sites.

    Role: Research Engineer

    Prediction of Protein-Peptide Binding Sites

    To develop a predictor which utilizes sequential and structural properties of a protein sequence to effectively train a machine learning model. To pick the best combination of features, genetic algorithm was used.

    Role: Research Engineer

    AARS - Automatic Attendance Recording System

    An application to record the attendance automatically in a class. We used android-studio to host and react-native framework to control the application.

    Role: Application Developer

    Phishing Detection using Machine Learning

    We performed detection of phishing websites through machine learning. We used the WEKA tool to clean the dataset and attribute selection.

    Role: Machine Learning Engineer

    Replication of Arm Movements using Computer Vision

    To model a 4DOF mechanical arm to replicate human movements. Used a Kinect camera to resolve the video into skeletal form. Solved inverse kinematic problems for a model of arm with 4 directions of freedom. Passed these calculated angles to the model for replication.

    Role: Programmer

    Education

    The University of New Orleans

    Master of Science
    Computer Science

    GPA: 3.75

    August 2017 - May 2019

    Smt. Kashibai Navale College of Engineering

    Bachelor of Engineering
    Electronics and Telecommunications

    Percentage: 68%

    August 2013 - May 2017

    Skills

    Programming Languages & Tools
    • Eclipse
    • Scikit-Learn
    • Rasberry-Pi
    • Arduino

    Interests

    Apart from being a machine-learning engineer, I enjoy most of my time being outdoors. I enjoy hiking and exploring new places.

    When forced indoors, I follow a number of sci-fi and fantasy genre movies and television shows, and I spend a large amount of my free time exploring the latest technology advancements in the machine learning world.

    Awards & Certifications

    • Graduate Research Assistantship
    • Privateer Scholarship Award
    • Neural Networks and Deep Learning - Coursera
    • Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization - Coursera
    • Structuring Machine Learning Projects - Coursera
    • 1st Place - Sinhgad Instiitutes, Pune - Sociobotics 2016
    • 2nd Place - Veermata Jijabai Technical Institute, Mumbai - VJTI Robotics 2015