Richa Patel

Boston, MA adyspe@r.postjobfree.com 857-***-**** linkedin.com/in/richapatel760/ Github Education

Northeastern University

Master of Science in Information Systems Major in Machine Learning Sep 2022 – May 2024 Boston, MA

Relevant Coursework: Probability & Statistics, Machine Learning, NLP, Data Mining, Data Visualization & Computation, Data Management, Deep Learning

Pandit Deendayal Energy University

Bachelor of Technology in Software Engineering

Aug 2018 – May 2022 Ahmedabad, India

Relevant Coursework: Creating and Maintaining Scalable Design Systems,cloud computing,web development,Data Structures,Java Enterprise Development with the Spring Framework

Skills

Programming Languages (Convolutional Neural Networks (CNN),PyTorch,Python,NumPy,Pandas), Developer Tools (Microsoft PowerBI, Tableau, Visual Studio Code, Microsoft Office Suite,TensorFlow,Scikit-Learn,PyTorch), Framework & Technologies (Pandas, Numpy, Deeplyr, Tensorflow, Amazon Web Services), Certifications (AWS Certified Solutions Architect – Associate) Professional Experience

Kaushalam Digital Pvt. Ltd.

NLP (Natural Language Processing) Engineer

Jan 2022 – Aug 2022 Boston, MA

•Designed a machine learning model to predict the churn rate among 10K customers based on historical data; reduced churn rate by 71% in the first week of implementation.

•Trained an encoder/decoder grammar error correction model using data methods; outperformed open-source baseline by 92%.

•Initiated a character-based transformer spelling correction model using Triton,reduced inference latencies below 178ms at 35 req/s.

•Developed an NLP system that automatically classified 7 5K emails as spam or advertising mail using 11+ natural language processing methods.

LogicRays Technologies Pvt. Ltd.

Graduate Engineer Trainee

May 2021 – Aug 2021 Ahmedabad, India

•Decoupled applications via queue design pattern for efficient processing of draft and Mastercard requests

•Implemented support for concurrent processing using multi-threading via Executor’s framework in java to improve system response time in EU regions

•Fixed security bugs of cross-site scripting and SQL injection in applications to maintain security compliance

•Collaborated with 5 business departments to develop 20+ key metrics by building interactive dashboards for stakeholders using PowerBI which helped to achieve their quarterly KPIs and identified their KRIs Relevant Projects

Predict closed questions on Stack Overflow

NLP,Scikit-Learn,Statistic

Jan 2023 – May 2023

•Built a predictive model using NLP techniques to classify Stack Overflow questions as closed or open.

•Developed a user-friendly web application that allowed users to input their questions and receive real-time predictions on question closures.

•Collected a large dataset of Stack Overflow questions and performed data preprocessing tasks, including cleaning and normalization.

•Extracted features using techniques such as bag-of-words and TF-IDF to represent textual data numerically.

•Employed machine learning algorithms, such as logistic regression and random forests, for model training and fine-tuned hyperparameters for improved accuracy.

•Evaluated the trained model's performance using metrics such as accuracy, precision, recall, and F1-score. Cascading Flight Delay Analysis

Python, Scikit, Seaborn, Weather API

Sep 2022 – Oct 2022

•Analyzed factors leading to cascading delay in departure of passenger aircraft such as weather, aircraft specifications, and incidents along with various airport features like size, and capacity after scrapping 2016 - 2020 data with over 10 million records

•Generated classification and regression models with 92% accuracy while predicting flight delay in minutes using selected features Information Retrieval Projects

Python, NumPy, BeautifulSoup, Scikit-Learn, Elasticsearch, NLTK, spaCy May 2021 – Jul 2021

•Implemented end-to-end pipeline to Index 86000 documents on Elasticsearch using Bulk API and ranked them using IR ranking models along with using Google’s PageRank and HITS to rank and rate pages respectively

•Implemented ML models for document ranking and evaluated results using trec_eval, precision, recall, f1-score and accuracy