Ansh Nirav Mehta

i Profile ć ad4d93@r.postjobfree.com ] linkedin.com/in/mehtanansh a github.com/mehtanansh Education

SUNY - Stony Brook University Aug. 2022 - Present

M.S. in Computer Science Concentration in Data Science GPA: 3.7/4 Courses: Computer Networks, Databases and Big Data, Data Science, Probability & Statistics, Visualization, ML Mumbai University K. J. Somaiya college of Engineering (KJSCE) Aug. 2018 - May 2022 B.Tech. in Information Technology GPA: 9.04/10

Skills

Languages & Databases: Python, C, R, Golang, SQL, JavaScript, MongoDB, C++, PHP, HTML Frameworks & Tools: Django, Flask, ReactJS, AngularJS, D3.js, Flutter, StreamLit, JIRA, Git, LaTeX, Heroku Additional Skills: Tableau, Power BI, Excel, SageMaker, TCP/IP, HTTPS, HTTP3, QUIC, SPDY, BGP, OSPF, ISIS, Web Scraping, Hypothesis Testing, Machine Learning, Computer Vision, Deep Learning, Natural Language Processing, Azure ML Studio, Data Wrangling, AWS, MapReduce, Big Data, Hadoop, PySpark, Polars, PyTorch, and TensorFlow. Experience

Entrigna June. 2023 - Aug. 2023

AI/ML Intern StreamLit, AWS SageMaker, Computer Vision, Tensorflow, Python

• Achieved an accuracy of 72.35% while developing a ResNet based the Discriminator of a Generative Adversarial Network model. Engineered Deep Learning systems, including their frontend and database with the help of Streamlit, AWS Buckets, Python and the company’s environment. Pred Healthcare Solutions April 2022 to May 2022

Backend Developer Intern AngularJS, Django, HTML, SQL, CSS, JavaScript, Gitlab

• Identified important KPIs and developed a dashboard for monitoring product usage of the healthcare information man- agement system. This functionality led to reducing the system failure reports from userś side by 79%.

• Revamped and enhanced the PDF Generation module which was used by 64% of the doctors during medical report generation and Prescription generation. Actively took part in the conceptualization of the Hospital inventory module. Intertrust Group June 2021 to August 2021

Python Developer Intern scikit-learn, Pandas, MongoDB, Selenium, Python

• Improved an existing ML model that was used for Financial Reconciliation. Reduced the lines of code from 3400 to 2960 which helped in reducing the latency resulting in improved performance.

• Wrote unit and service level tests using Selenium for the services for the application leading to a decrease of 14 bug reports related to the UI.

Software Development Cell - KJSCE Jan 2020 to March 2020 Web Developer Intern PHP, Firebase, Bootstrap, HTML

• Constructed an engaging website for personality trait identification utilizing the OCEAN model through a question and answer system. Gathered and analyzed data from 600 Engineering Department students, revealing common personality traits of Openness and Extroversion.

Projects

AirBnB Property Analysis for Customers l March 2023 - May 2023 Visualisation of Airbnb Listings across NYC Flask, D3.js, JavaScript, Python, HTML, CSS, Bootstrap

• Designed and implemented an interactive AirBnB Dashboard that that offers streamlined house search options, includ- ing filters for price, location, lease type, house types. It comprised of maps, box plots, bar plots, and other intricate visualizations to display user-preferred house listings in each borough and neighborhood, optimizing the experience. Healthino June 2021 - May 2022

Disease Prediction and Chatbot for Mental Healthcare Python, Django, TensorFlow, ML Libraries, JSON

• Guided a team of four to achieve a 98.6% accuracy in the Disease Prediction module by constructing an Artificial Neural Network (ANN) model. Implemented LSTMs, Bidirectional LSTMs, and Sentiment Analysis, attaining an 83.4% accuracy for the chat-bot module. Published the research outcomes in journals including Springer & IEEE. Game-Dise Jan 2021 - May 2021

Web App for playing Sudoku and Tic Tac Toe Flutter, Dart, Firebase, Heroku

• Led a group of 5 and implemented a Flutter-based sudoku and Tic Tac Toe puzzle game. Implemented principals of AI-based Game playing, Min-Max Algorithm and Back-tracking on Dart for various user-modes and difficulties.