Evan Downs Email & Phone Number

Kingsport, TN, US

Kingsport, Tennessee, United States

- Constructed models across different manufacturing process lines for predicting quality metrics- Developed scalable industrial data pipelines from historian's recorded data to ML model ingestion- Trained 5 engineers in best practices in applying data science to industrial data- Designed simulations and optimizations for recommending set points in.

Kingsport, Tennessee, United States

• Constructed a multi-model pipeline scanning 944 different product datasets to provide improvement opportunity rankings
• Developed a Streamlit application displaying improvement opportunity to guide stakeholders in interpreting insights
• Utilized distribution fitting in market size estimation guiding a business’ production plan for a new product
• Optimized topic model selection in a Latent Dirichlet Allocation model for a self-service text analytics application

Atlanta, Georgia, United States

• Led a multi-disciplinary group of 6 engineers in building a home energy management system
• Built an Azure hosted web application comprising of 5 custom built Node.js APIs for data streaming, a login page with basic authentication, a React.js energy statistics table, and 4 D3.js live and historical data.
• Optimized Azure SQL Database comprising of 6 tables for handling 86,400 database entries and 57,000 queries per day
• Developed 10 sentence summary of 75 customer interview responses using TensorFlow’s Universal Sentence Encoder
• Web scraped and preprocessed 300 critical reviews of competitors’ products to gain problem landscape insights.

Spartanburg, South Carolina, United States

• Developed a vehicle flow sankey visualization tool in D3.js that identifies bottlenecks in real time by measuring accumulation and speed of vehicles and comparing them to their respective averages
• Created a late vehicle detection algorithm built on a logistic regression model that is able to flag when a vehicle will be late with 92% accuracy, 93% precision, and 91% recall. The algorithm can also assign blame to.
• Performed a study on vehicle flow patterns using created algorithms and software visualizations that identifies the root cause for why vehicles are late and provides recommendations on how to solve chronic issues for.

Spartanburg, SC

• Worked on software development, machine learning applications, data analytics, and database management
• Developed a data quality web application that displays both live and history errors in defect and rework entry processes. Has saved the company $4M since its implementation
• Constructed an electrical defect heat map web application that uses a K-means machine learning algorithm built with Tensorflow.js to form clusters of defects on wiring harness images. The heat map is displayed using.
• Constructed a recurrent neural network with long short term memory to flag anomalous vehicle patterns

Spartanburg, SC

• Tasks included software development focusing in web application development through Oracle Application Express
• Developed 8 web applications focusing on data visualization and data analysis of defect trends
• Built image storage web application for uploading and viewing defects on a car
• Gave daily reports on rework performance and maintained a problem list for a new series launch

Atlanta, GA

• Research consisted of discovering how machine learning algorithms could be used to describe non-linear differentialequations.
• Developed a Neural Network Algorithm to find the final trajectories in a Lorenz Attractor
• Completed an 11-week Stanford Machine Learning class on Coursera.

Aiken, South Carolina

Worked in Site Infrastructure Engineering developing work packages for installation of transfer switches, pumps, and motors, writing engineering evaluations for steam and HVAC systems, coordinating soil tests, and formatting technical writing documents.

Atlanta, Georgia

• Advised and instructed students in a physics laboratory setting, prepared lab equipment, prepared lesson plans, and gradedpapers.
• Advised a total of 23 labs and instructed 60 students.

Master'S Degree, Analytics, 4.0/4.0

Bachelor'S Degree, Mechanical Engineering, 3.94/4.0

Activities and Societies: • Tau Beta Pi Engineering Honors Society • Pi Tau Sigma Mechanical Engineering Honors Society, Director of.