Nathan Frank work email
- Valid
- Valid
Nathan Frank personal email
Nathan Frank phone numbers
Skilled data scientist with experience leading all aspects of the machine learning life cycle. Former Astrophysicist turned full stack data scientist with proven history of delivering results into production while leading projects with international, cross-functional teams. Versatile problem solver and intermediary between data science and engineering with an intuition for building machine learning technologies.
-
Director, Machine Learning Operations & PlatformGrainger Apr 2023 - PresentLake Forest, Illinois, Us -
Director Of Machine Learning EngineeringStrong Analytics Apr 2021 - Apr 2023Chicago, Il, Us -
Senior Data ScientistStrong Analytics Aug 2020 - Apr 2021Chicago, Il, Us
-
Ai Scientist [Tech Lead, Ml Platform]Stats Perform Jul 2019 - Jul 2020London, England, Gb• Led the development of a machine learning training and deployment framework, built on Docker, EKS and SageMaker, to dramatically speed up iteration cycles and reduce time to deployment. Trained teams internationally on using this and other technologies in their workflow.• Spearhead development of ML Platform roadmap to modernize company’s AI infrastructure and improve efficiency; evaluated potential tools (AWS Sagemaker Studio, ML Flow and Kubeflow). Advised on team resourcing and project allocation.• Trained and deployed model using GNNs to predict minutes played of NBA players with RMSE (5.75) lower than human domain experts (6.15). -
Ai Scientist [Tech Lead, Stats Vq]Stats Perform Jan 2019 - Jul 2019London, England, Gb• Led international team of 13 data scientists and engineers in building a predictive player props API for sportsbook customers, generating $500k of new business in the first six months. Oversaw on‐time completion of deadlines for the NBA & NFL 2018‐19 playoffs as well as 2019‐20 season, requiring delivery of 11 new models into production while automating existing manual process for legacy products.• Drove development of an internal TensorFlow based model development framework with libraries for dataset generation, versioning and metadata management, architecture definition and training, and deployment to a TensorFlow Serving RESTful API in AWS Fargate.• Provided technical oversight of the VQ product: Java Spark ETL pipeline, model development, cloud infrastructure(AWS), human‐in‐the‐ loop interface and both internal and client‐facing APIs. Guide story definition and refinement; led weekly/quarterly planning sessions.• Aided design and creation of cloud data lake (AWSS3, Glue & Athena), reducing cost and easing access to historical data archived in on premise Oracle databases. Develop methods to query the data lake, process and split the result into datasets and store in S3.
-
Data ScientistStats Perform Jan 2017 - Jan 2019London, England, Gb• STATS Prediction as a Service (PaaS): Co‐authored tools integrating ETL pipelines to deliver data to the cloud (AWS S3), generate datasets from queries executed at scale (AWS Athena), train TensorFlow models in Docker and deploy to a HTTP RESTful API endpoint (TensorFlow Serving with AWS SageMaker, Lambda & API Gateway). The design won the inaugural VCG AI/ML Hackathon most disruptive entry.• STATSEdge: Refactored legacy data science pipeline and deployed existing scikit‐learn models to Flask APIs. Contributed to Apache Spark ETL pipeline in a paired programming (XP) setting. Optimized existing models, reducing runtime by an order of magnitude.• NBA Live Win Probability: Achieved SOTA accuracy (88%) predicting end of game outcomes for NBA games given in‐game context through the use of embedding techniques. Processed and released a curated dataset containing 352 features of over 8.7M NBA plays from the 2002‐03 through 2016‐17 seasons. Presented work at 2018 Sloan Sports Analytics Conference.• STATSInsights: Used LSI, LDA and other topic modeling techniques to classify ∼80k human generated media notes, learning templates to automate nearly half (44% ). Delivered 546 automated notes using custom Jinja‐based templates for Allstate’s March Mayhem campaign during the 2017 NCAA Tournament. Used learned templates to generate >19k notes across 10 international soccer leagues.
-
Research/Teaching AssistantUniversity Of North Carolina At Chapel Hill Jul 2011 - Jan 2017Chapel Hill, Nc, Us• Developed an analytic model of afterflow emission from structured, off‐axis Gamma‐ray Bursts (GRBs) with results comparable to 3D magnetohydrodynamic simulations. Modeled spatiotemporal observational data using a genetic algorithm based optimization algorithm.• Builder of PROMPT-SSO and contributor to the Skynet Robotic Telescope Network: a collection of fully automated telescopes, control software, image and data processing pipeline, and web interface.• Head instructor for Introductory Astronomy Lab. Organizer and host of Morehead Observatory Guest Night, a weekly public astronomy presentation. Instructor for ERIRA, a summer astronomy research and field experience. -
Accounts Receivable ManagerPatzik, Frank & Samotny Ltd. 2009 - 2011Chicago, Illinois, Us
Nathan Frank Skills
Nathan Frank Education Details
-
University Of North Carolina At Chapel HillPhysics & Astronomy -
University Of North Carolina At Chapel HillPhysics & Astronomy
-
University Of California, Santa CruzPhysics (Astrophysics)
Frequently Asked Questions about Nathan Frank
What company does Nathan Frank work for?
Nathan Frank works for Grainger
What is Nathan Frank's role at the current company?
Nathan Frank's current role is Director, ML Ops & Platform @ Grainger.
What is Nathan Frank's email address?
Nathan Frank's email address is na****@****unc.edu
What is Nathan Frank's direct phone number?
Nathan Frank's direct phone number is +191996*****
What schools did Nathan Frank attend?
Nathan Frank attended University Of North Carolina At Chapel Hill, University Of North Carolina At Chapel Hill, University Of California, Santa Cruz.
What skills is Nathan Frank known for?
Nathan Frank has skills like Statistical Modeling, Data Analysis, Python, Research, Statistics, Astrophysics, Quantitative Research, Bayesian Methods, Latex, C++, Scientific Writing, Microsoft Office.
Free Chrome Extension
Find emails, phones & company data instantly
Aero Online
Your AI prospecting assistant
Select data to include:
0 records × $0.02 per record
Download 750 million emails and 100 million phone numbers
Access emails and phone numbers of over 750 million business users. Instantly download verified profiles using 20+ filters, including location, job title, company, function, and industry.
Start your free trial