Drove data strategy for Woven's Benchmarking program where we compared our Advanced Driver Assistance Systems (ADAS) to those of our competitors, to inform and advise top leadership on business decisions. Created all data pipelines and transformations to deliver key insights to our organization via monthly business reviews and presentations to stakeholders and management. Created essential analytics and reporting for Woven's cloud costs that resulted in a monthly savings of $2 Million from Woven's monthly peak, among additional uncovered and gained efficiencies.Supported and improved critical legacy data pipelines from Toyota's acquisition of Lyft L5 to deliver Autonomous driving data at scale, via BigQuery and Flyte ETL to downstream tools and cross-functional departments.

Support the organization's B2B online marketplace by implementing a language agnostic data framework to rapidly deploy data applications using Docker, CRON, and AWS ECR.Maintained our Snowflake data warehouse and all warehouse functionality: data lake support, storage integration configuration, compute resource allocation, automated view building, and role permissioning.Maintained all ETL pipelines into the data warehouse using different ETL Tooling: Stitch data for the various tools we used(Salesforce, Sendgrid, etc) and AWS DMS for change data capture replication of production Postgres services.Created a recommendation engine using alternating least squares to recommend products to users based on their transfer history within a state’s seed to sale tracking system (Metrc).Created a supply chain network data set based on a state’s seed to sale tracking system (Metrc) that could allow a supplier to see current stock levels downstream of products they have sold to their customers. This data set is now a core subscription product offering called Relationships at Confident Cannabis.Created an abstract data pipeline that allowed the data team to deliver complex data objects to our production Postgres environments. This in turn allowed the backend team to build read-only Django models on top of the data objects we deployed for rapid product development.Supported the sales team by creating intelligent ETLs into Salesforce to attribute internal data insights to Salesforce related objects (Opportunities, Accounts, Events, Custom Objects, etc).Built out custom dashboard reporting within our app using Django to build Plotly JSON chart objects that could be rendered on the front end.

Managed a team of seven data scientists and analysts that solved problems across different teams at Shipt: Search, Product Catalog, Shopper Application, and Member Application. Managed projects include: Member autopilot recommendations (reordering on an automated schedule), shopper out of stock recommendations, and shopper pick order recommendations.

Built and deployed a convolutional neural network model that predicted category taxonomies for products missing categorical information that improved having to manually assign taxonomies before being browsable in the application.Designed a new offering system to match shopper skill with order difficulty using ELO principals.Onboarded, trained, and mentored seven data team members; setting them up for success at Shipt.Developed internal tools and libraries to improve workstreams for others on the data team to deploy data products and build ETL pipelines.Maintained and improved the data team's dedicated databases (Redshift/Postgres): vacuuming policies, resource management, automated view building, and setting schema structure.Created the definition for user search performance and leveraged the definition during numerous A/B testing as we migrated our search service from Algolia to Elasticsearch.Worked cross functionally with engineers on event instrumentation, tracking and data governance to ensure appropriate signals were measured across the application.

Managed data science tasks across our analytics team that controlled automation, machine learning, and data warehousing.Created unique data visualization tools using Tableau and Shiny Applications that turned data into insights.Wrote complex queries, stored procedures, and ETL logic in MySQL/SQL Server to track user behavior across Vinyl’s web technologies.Developed proprietary machine learning models that tackled a wide variety of vital business issues: fraud detection, call contact scoring, and lead quality scoring.Developed a robust recommendation engine using collaborative filtering techniques that decides which products to show based on the monetization per conversion.Used constrained optimization techniques that mapped users to an optimal area code to call from based on the lead’s geolocation, increasing pickup rates.Experience with marketing channel analytics for call center, paid search, email, and business development in the lead-generation space .Communicated and presented complex data insights in a clear and concise manner to a diverse audience.

Teaching Assistant for Probability, Statistics, Calculus I, and Calculus II.

Managed a team of up to 12 people where I created a cohesive work environment that made the business successful.