Summary
Joel brings almost a
decade of experience leveraging data & analytics to drive business growth and improve
operational inefficiencies. Joel has led teams of scientists & engineers, built end-to-end machine learning
pipelines and developed algorithms to improve marketing and sales effectiveness -
one classification model
significantly reduced customer churn likelihood, increasing average revenue per sales representative by
~$400k. More recently at Brightloom, Joel developed a generalizable machine learning application and experimentation platform,
offering data science-as-a-service to major fast-food and fast-casual restaurants. Additional projects include a
logistics reporting solution at Lululemon, optimized marketing campaign spend by channel for Funko, a recommendation
engine for the Pursuit Collection, and data governance policies and procedures for Bluetooth SIG.
Core Tech
Programming / Scripting:
- Python: pandas, numpy, pandapy, dask, glob, scikit, hpf, seaborn, matplotlib, etc.
- SQL, MySQL, PostgreSQL
- DAX, Power Query, MDX
- R: limited working knowledge. efforts of late have been focused on deepening python understanding
Tools:
- Azure Suite: cosmos, cognitive services, machine learning studio, functions, devops, catalog, data lake
storage, blob, etc.
- AWS Suite: sagemaker, glue, s3, redshift, ec2
- Model Management: databricks, ml flow
- Report Development: power bi, tableau
Project Delivery:
- Jira, Confluence, Git/GitLab/Bitbucket, Airtable, Smartsheets, et al
Primary Industries:
- Retail: sales / demand forecasting, recommendation engine development, supply / inventory management, cltv
prediction, customer segmentation
- Finance: risk assessment, timeseries forecasting, dynamic portfolio management, valuation modeling
- Logistics: route optimization, pricing strategy, resource strategy
- Marketing: cac optimization, content strategy