Devansh Gupta Email & Phone Number

Dallas, TX, US

Columbus, Indiana, US

Columbus, Indiana, US

- Part of the Perception unit at Advance Research and Development team.- Contribute in creating Fully Autonomous Forklifts. - Work on creating vision based Deep learning models converting to the ONNX models and TensorRT engine for inference.

Cleveland, OH, US

• I will be working under Dr. Stephanie Griggs on Young adults with Type 1 Diabetes and Sleep Deprivation during COVID 19.
• Engineered the wavelet analysis model used to find the correlation between glucose data, activity patterns, and sleep deprivation in Type 1 Diabetic patients in Python and MATLAB.
• Designed an in-house algorithm for time-bands creation to simplify the data analysis in C++ and used OpenGL for 3D visualizations.
• Applied various machine learning techniques by deploying Python to build time series analysis visa low-rank matrix factorization of the sleep patterns.

Uniontown, Ohio, US

• Created analytical datasets by extracting, manipulating, and consolidating huge datasets with >4 million records using Mongo Client in Python for FedEx warehouse by mapping MySQL to MongoDB.
• Conducted regression tests on the data to validate the Weighted Average Cost of Capital (WACC) and its implications for Incentive Plan Design for the FedEx drivers.
• Analyzed the 16.8% performance increase in driver's Lease activation ROI performance based on online training versus in-person training using Alteryx on Docker system with a dataset of 500k drivers.
• Conducted a data regression analysis of the relationship between company stock prices and industry trends, achieving a 15% more accurate prediction of performance than in previous years.
• Utilized web scraping techniques to extract and organize competitor data.
• Updated company data warehousing techniques such as data recall and segmentation, resulting in a 20% increase in usability for non-technical staff members.

Cleveland, OH, US

Hyderabad, Telangana, IN

• Conceptualized & implemented an Aspect based sentiment analysis dashboard used by Amazon India using AWS Cloud9 IDE, data management using Amazon S3 Console and data visualization using AWS QuickSight.
• K-NN model development, search ranking & validation of 100+ financial products and services using gathered data to rank the reviews & evaluate sentiment using POS tagging via NLTK (Stanford NLP).
• Developed a Flask API to understand customer behavior leading to 95% success in aspect words generation.
• Video and image processing using GPU parallel computing, and stream analysis with audio/video codecs like MPEG, AVI using ffdshow by creating a Speech-to-Text API synchronous recognition request.
• Developed parallel image/video processing algorithms in depth image-based rendering (DIBR), scale-invariant feature transform (SIFT), and region labeling for E-commerce reviews.
• Optimized these GPU-based real-time data analyses and speed up more than 10 times.

• Developed Knowledge Extraction pipeline for public documents with Semantic Extraction using NER, Constituent Parsing with Rasa NLU & Spacy and built a Knowledge Graph using Dgraph, GraphQL to represent the extracted.
• Maintaining, training and deployed text classifiers API with Word Embeddings, LSTM, BERT using TensorFlow, Keras on AWS Sagemaker.

IN

Master'S Degree

Computer Science, Computer Science

12Th