Jason Roell
Head Of Engineering at Vurvey
Cincinnati, Ohio, United States
Summary
👤
Senior🎓
Top SchoolJason Roell is a Head of Engineering based in Cincinnati with over a decade of hands-on experience building cloud-native, high-throughput systems and leading cross-disciplinary teams. He has designed architectures and CI/CD infrastructure for startups and enterprises alike—shifting monoliths to microservices at Kroger, driving AWS and container strategies as a CTO, and launching a production video-survey pipeline at Vurvey that handles encoding, transcription, and translation in Kubernetes. Equally fluent in product and technical roles, he has translated customer needs into roadmap decisions as a Technical Product Manager and founded a data-science consultancy delivering ML and deep-learning solutions. Known for squeezing large performance gains from legacy systems (one project improved throughput by 900%), he blends pragmatic engineering, mentoring, and hiring to scale teams and systems. Jason maintains a public presence through jasonroell.com, reflecting a curiosity for applied ML and production-grade infrastructure.
10 years of coding experience
13 years of employment as a software developer
Bachelor's degree Computer Information Systems, Bachelor's degree Computer Information Systems at University of Cincinnati
Stackoverflow
6,739reputation
751kreached
107answers
18questions
Github Skills (28)
recurrent10
predict10
market-prices10
stock9
recurrent-neural-networks9
prediction9
parameter-tuning8
financial8
sam7
hyper7
asp-net6
deep-learning6
task-parallel-library6
entity-framework6
sql6
Programming languages (7)
Github contributions (5)
jroell/StockMarketPredictionsWithRNN
Jul 2017 - Sep 2022
The goal of this project was to predict stock market prices using a recurrent neural network. This project was inspired by the Standford paper "Financial Market Time Series Prediction with Recurrent Neural Networks Armando Bernal, Sam Fok, Rohit Pidaparthi December 14, 2012" . Using the Standford paper as a baseline, I was able to improve and beat the accuracy of their model using hyper parameter tuning of my RNN network.
Contributions:1 review, 2 commits, 1 PR in 5 years 2 months
goalpredictstockbeattime-series
jroell/RecommendationEngineWithDeepAutoEncoders
Jul 2017 - Feb 2023
The goal of this project is to build a recommendation system using Deep Auto Encoders
Contributions:2 pushes, 1 branch in 5 years 8 months
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.