David Grover Email and Phone Number

As a senior leader I drove strategic improvements to our consulting practice, initiating architectural runway projects across our data engineering, data science, data modeling, visualization, software engineering and project management practices. This dropped platform deployment time from three months to two weeks, data modeling effort from months to weeks, standardized data architecture across GCP, AWS and Azure, and led to significant improvements in employee morale, retention and productivity. I contributed to projects as a chief architect, as a client lead working with senior executives at clients, as a project lead managing delivery teams of data engineers, data modelers and data scientists, and sometimes as an ad-hoc resource providing data architecture and data modeling expertise as needed. Clients ranged from oncology medical devices through retail commerce to government.Other value add activities were varied. In my first year I developed employee resource groups in mentorship, collateral and DEI to make the company more resilient and responsive. Most recently I formalized our case study collection process, collected materials for business development efforts, and wrote more than 40 case studies covering more than ten years of company projects.Technologies were primarily Snowflake, Airflow, Tableau, BigQuery, S3, ADF and ABS.

Raybeam was acquired by DEPT in January 2002. My work for Raybeam is gathered under DEPT.

First, I developed a methodology to measure and improve the quality and efficiency of data management operations. The central concept of the methodology is “semantic debt,” which is a measure of integration cost. Notes and graphics are posted at www.semanticdebt.com, and articles and decks are published using the LinkedIn publishing tools.Second, I wrote a practical, cookbook-style handbook for data management teams. The book provides a simple set of instructions for building the major data management systems, including the Data Lake, the ODS and the Kimball Star Schema, and a set of organizational expectations for each of those systems. The book provides a very straightforward way to design a data management strategic plan.

First, we performed an agile data modeling exercise to support a knowledge base integrating MDM to curate research for genetic variant/disease/drug combinations. I also attempted to clarify roles and key questions around the software architecture of the product. Second, I constructed a strategy to build a transformative system to integrate scattered public and private corporate research data, to handle GDPR requirements. Execution on the strategy required building decks that could sell the vision to internal audiences. Part of my time was also spent building an internal collaboration framework for data scientists and data managers, producing in a short time the first steps of a data governance mechanism to support the new system. I also extracted a bunch of publicly-available TCGA data and loaded it into a prototype data lake using Python, keying the data to cross-reference it with clinical/patient attributes. Finally I built some demonstrations in Tableau. This was the simplest part of the architecture but would have provided enormous value as a demonstrator.Tools included Oracle, Python and Tableau.

Initially hired to mentor, manage and expand a newly-formed BI team at a financial services company, my role was to inject experience into the BI function to help it mature operations and strategy. After six months I transitioned into a Data Architect role, where I built a strategic customer-interaction warehouse and architected and drove the organization of email operations, Adobe Analytics implementation, and a customer MDM system.Tools included Netezza and Tableau.

Drove data management strategy at a multinational health-care company specializing in oncology devices and practice management software. The company needed an integrated approach their data management practice, from data modeling in OLTP systems to data warehousing, data engineering, ODS and MDM design., such as a Cancer Registry product that functioned as an MDM system. I also managed development of a new ODS and led the team building out the phase I ODS.Tools included SQL Server, SSIS and Birst.

Managed a business-oriented data management team that also performed program management and business analysis, in a very active database marketing group. The group was in the process of migrating from a ten-year-old fragmented email marketing database spread across ten different subsystems to a more sophisticated data management system combining various “big data” sources with more traditional customer engagement points and references. I was hired to architect and manage the development team. Much of the technical work was transferred to IT early in my role as part of a strategic realignment, and the role became a more traditional marketing operations function. I also prototyped a clickstream warehouse for SiteCatalyst data and various analytics processes for email management and marketshare, formalized the Privacy management function, and processes for establishing boundaries in customer engagement. Tools included Oracle, SQL Server, SSIS, Pentaho and Hadoop/Hive.

Designed and implemented a data warehouse sourced from the Atlas Public Health application database, a mixed document/event-model system used to track public health cases. Developed reporting strategy to drive migration from Excel and email. The strategy was implemented with MS Reporting Services 2008, using report models, parameterized suites and SSAS.

Advised Canadian IT management at a financial services company on business intelligence strategy, analytics, data warehousing and data management, and prototyped and developed ETL, dashboards, reporting suites and ad-hoc tools. I also provided ERP migration expertise for a long-overdue Axapta upgrade. Infrastructure was based on SQL Server 2005, Axapta, Dynamics, SSRS and in-house applications.

Hired and motivated a team of data warehousing and BI professionals to manage the data warehouse infrastructure at an ecommerce company. We automated and expanded reporting for various Finance, Engineering and Operations groups, and drove the implementation of end-to-end high-volume customer experience reporting across error-logging, Gomez, clickstream, email marketing, quality surveys and Tealeaf data, which gave Engineering and Marketing their first hard insights into customer and site behavior. I also led initiatives to formalize transaction database development, change management, database operations interaction model, and database design processes. Infrastructure was based on SQL Server 2005, Informatica, Pentaho, Microstrategy and SSRS.

Conceived, implemented and rapidly expanded the BI organization for the Supply Chain group at Intuit, which eventually became the operational BI solution for much of the company. The system included the full breadth of data management infrastructure, from ETL out of thirty data sources, MDM, quality auditing, an ODS and a warehouse. My team functioned as data product managers and we integrated closely with the operating mechanisms of our customers. We also built applications to automate reference data collection, which added extraordinary value to the base data. Some of these applications managed very complex business logic, including Monte-Carlo simulations and manufacturing planning.Prior to working in Supply Chain in 11/2002 I built an integrated customer-service data warehouse, sales warehouse and clickstream-traffic warehouse for their ecommerce group, and a surprisingly useful and resilient CRM system for a second-level CSR group.Infrastructure was based on SQL Server 2000, DTS, SSRS and some Business Objects, against Oracle 10.7 and 11i, Seibel, legacy mainframe applications and various FTP and third-party sources. Ecommerce warehouses were built against ATG, Seibel and home-grown apps.

Brought back to rearchitect the ETL and design of the data warehouse I'd built the first time I was at Beyond, to accommodate a super-spiffy stateless ecommerce system built during the interim. This warehouse would eventually have been bundled as an option for users of the ecommerce system. I also expanded the existing warehouse infrastructure to manage clickstream data from a Teradata source. Second, I was challenged to improve efficiency in the fraud department. I prototyped a machine-learning (ML) system based on a genetic algorithm, using Bayes Theorem and recursive ETL to model the existing fraud-detection processes, and implemented pieces of the system in stages with fraud personnel to provide continually-improving ROI.GA development was on SQL Server. Warehouse development was primarily on Sybase using DTS against Oracle and Teradata.

Data Architecture and team management at a busy small web-application consultancy. We built everything from pretty brochureware sites with customization capabilities to complex combined incident management/account management/content management applications with integrated custom reporting and warehouses. Also the occasional ecommerce site. All the work was done in small teams in close contact with the client, using what's now called an "agile" methodology. Most of the database management was done in SQL Server, and my team was hired for SQL Server experience.

Tasked with fixing reporting at an online software retailer/distributor, I first analyzed and formalized reporting requirements for internal and external customers and then designed and implemented, in short scrum-like iterations, a data warehouse and simple control-table based reporting application. ETL was exclusively out of a legacy ecommerce transaction system based in Sybase. The reporting app was built quick-and-dirty in Access, and management had to physically delete the app four months later before people would use the new Business Objects toolset. I also prototyped a market-basket analysis system for Marketing, did QA on Informatica transformations for the long-delayed enterprise warehouse, and provided business-logic expertise for an SAP implementation.

Management asked me to figure out the technical infrastructure for a key product: a way to assign unique UPC-like codes to items across foodservice (or any vertical) catalogs, what's now called a SaaS MDM product but which was a brand-new concept at the time. So I researched, designed, developed and implemented through prototype phase to production a multi-domain data normalization process, my first ever. I helped hire, build and train the team to run the resulting data management system, which served as the infrastructure for a suite of profitable information services for many years after I left. I also developed demo and production data warehouses for knowledge management, reporting and data quality systems. In my first year at Instill I performed market research, competitive environment, ROI and pricing analyses and databases for corporate business plans.