Brad Eric Jacobson

Brad Jacobson 3319 Sweergum Cedar Park, TX 78613 - –- Career Objective To make material contributions to business with process design, data architecture, data modeling, database programming and statistical analysis. Qualifications Metadata management including table diagrams, data dictionaries, glossaries and solution data flow diagrams (Embarcadero ER/Studio & Sybase power designer), 10 years+ OLTP database design/implementation, insert/update/delete stored procedures to preclude SQL injection, specialized stored procedures, views and query optimization, 10 years+ Warehouse/data marts: ETL design and implementation, enterprise data model and dimensional modeling, utilizing SSIS and PL/SQL stored procedures, 9 Years + Project management and user requirements gathering, excellent oral and written communication skills. Comfortable working with business people to develop tools to arm them to participate in markets from their chosen market position, 20 years+ Data analytics including robust statistical modeling of consumer behavior with associated R-squared and P-value calculation in the database as well as integration with R, 4 years+ SQL tuning, SQL training and PL/SQL and TSQL database programming, 12 years+ Strategic contributions Solved a problem around re-positioning empty railroad cars with AI multi agent system Re-invented expectation maximization for brick and motor store visit modeling, R squared frequently 0.70 or better Developed a detailed recommendation for the sponsorship of a tokenized commercial real estate market Education BA CSC - Southwestern University in Georgetown TX Graduate work in CSC at UTA (University of Texas at Arlington) Professional Experience Fidelity (Via Hashmap) – Chicago, IL Title: Principal Software Engineer (Data Modeler – Kimball Method) Manager – Sunil Chilliyil Time Frame – August 2021 – Present I worked with several gate keepers to arrive at data warehouse design components that made everybody happy and rationalized and consolidated the data. Which is to say, the general inclination was to just restate the data from the source rather than do any rationalization and consolidation. I also participated some in reviews and setting standards; although in theory I was only hired as a data modeler. ReVyvv – Dallas, TX Title: Director of Data and Artificial Intelligence Manager – Effectively the project manger : Stephanie Kennedy Time Frame – August 2020 – July 2021 I designed and maintained the databases for the cloud and the phone to support a mobile e-commerce platform. I deigned and implemented the syncing process and wrote most of the C# end points that allow the app to communicate with the cloud database. I also did all the math that arose during implementation and worked with the Dean of mathematics at Kent state , one of the top mathematicians in the world, to develop a unique price optimization algorithm. Texas Capital Bank – Richardson, TX TitleLead Data Modeler (Architect) ManagerJason Lowe & Brian Coats Time FrameOctober 2016 – July 2020 (Including 5 month phase) My biggest contribution was on a conversion project where we moved from FIS loan system to ACBS loan system. I laid out a database design in ER/Studio using Kimball method, wrote the hardest SQL for the ETL programmers, helped the ETL programmers with SQL they wrote both fixing it and optimizing it, and wrote the hardest ETL as stored procedures for the biggest technical project ($20,000,000+) Texas Capital Bank, a 27 billion dollar commercial bank, has ever embarked on, called loan core. Please contact reference Jason Lowe listed below. I worked on data glossaries and data dictionaries as well the approach and data flows for he project along with the actual modeling. For example, I promoted and implemented the idea of conforming as much as possible to existing communication protocols with downstream systems as we changed over to the new loan system to minimize our cost and time to market. I also worked independently on data quality. For example, I brought to the business my findings that large loans and credit arrangements (loan commitments) that were funded by more than 1 bank (called syndications) were accounted for with more or less than 100% funding in some cases. I worked with the leaders of credit fulfillment and the other technical resources to solve many problems associated with integrating the new loan system into many other bank operational systems, in time becoming a SME in the our loan operations. The data for this project is hosted on the cloud in Atlas and Azure for staging and processed data respectively . Also, more recently, I worked on a smaller project solo for a profit center at the bank that loans out money to finance business insurance policy premiums during which I fulfilled a much broader role and again the data was hosted on the cloud in Atlas and Azure. The approach for the last project envisioned that we were to become full stack developers so I actually wrote some SSIS packages rather than jut helping the ETL programmers with their code and created one Power BI cube. Realpage – Richardson then Carrollton, TX TitleData Modeler/Data Architect ManagerJose Soto Time FrameFebruary 2016 – September 2016 Implemented a star scheme for their billing transactions – Kimball Method. I introduced better data integrity and performance than they have seen in their data warehouse. I implemented the data migration from the previously built transaction table to the star scheme. I also analyzed their invoice credits which are substantial in the hopes of reducing such credits. We used SQL Server and SSIS packages to do all of the ETL; design and data flow documentation were done in Visio. Modeled and implemented with an eye to Cognos needs, for example minimizing the use of snow flake scheme. The billing transaction log data mart included pre-built aggregations of the fact measures per quarter for 17 of the 40 dimensions. Improved data quality by introducing uniqueness constraints on the dimensions and foreign keys from the fact table to the dimensions. These were new to real page and an innovation I was introducing. ERL was performed with SSIS and stored procedures. Citibank – Irving ,TX TitleData Modeler ManagerPrasad Amati Time FrameAugust 2015 to February 2016 Implemented minor scheme changes to existing products making minor contributions to database integrity and data quality. Translated XML hierarchical model into relational representation. Worked with golden gate to hold schemes in sync across multiple environments. Used Erwin to document current scheme structures. We used Oracle and PL/SQL for the ETL, documentation of design and data flows was done in Visio. GM Financial Arlington, TX TitleData Architect ManagerChris Holk Time FrameFebruary 2015 to August 2015 Worked with business analyst to ensure approved changes fully reflected requirements. Took responsibility for shepherding change requests through the internal GM Financial process. Implemented minor scheme changes to existing products (Data Vault Method) making minor contributions to database integrity and data quality. Used Erwin to document database structures. Our DBMS platform was Oracle and ETL was done in Informatica. Belo Corporation (parent company of The Dallas Morning News) – Dallas, TX TitleData Modeler/Data Architect ManagerDavid Hill Time FrameOctober 2008 to January 2015 Designed and implemented a data warehouse as solo effort (I got some help late in the project). I designed the data extraction mechanisms, the canonical model, the ETL process and the dimensional models. I implemented the data migration process. I created a glossary and established rules of governance as I was granted additional staff. Although I never gained access to all the data that would be appropriate, I did create and support a consumer targeting program which brought BELO $3,500,000 in incremental revenue. Please contact business marketing resource customer Nate Breindel -. At Belo we used both Oracle and Postgres and the database design was developed and documented in SyBase power designer. The ETL was done with PL/SQL stored procedures hand programmed to respect operational presidents constraints and to be run in parallel on subsets of the data to improve performance. I designed and built the warehouse in a way to try and make it the most useful to Cognos documenting natural and foreign keys as well as minimizes the use of snow flake scheme. I worked hard to learn the true pain points and find the real opportunities for a data warehouse to make material contributions. The data warehouse included many pre-built aggregations of the fact and dimensional data. They were effectively materialized views that we did not tell oracle were materialized views and instead of getting oracle to populate them, since they were too complex for Oracle, we had our own ETL mechanism maintain them as a final step in the processing. Dimension table always had uniqueness constraints and fact tables always had foreign keys to the relevant dimensions. Texas Guaranteed Student Loans – Austin ,TX TitleData Architect ManagerRose Hitnze Time frame February 2001 to August 2008 Designed and implemented several databases, including a unified institution database and participated in major data migrations; performed data warehousing, SQL tuning. I wrote manual scrubbing routines to eliminate PCI data but conserve data integrity. I also wrote scripts to generate entirely fake data on occasion. Worked directly with function area leaders to help them with their real problems. On many projects I brought forward business process designs that were accepted by the technical and business stake holders. I view the data architect’s proper role as a business problem solver not a full bore technical geek, although we all like our technical geeks. We used some DB2 but mostly Oracle. ETL and other database centric processing were hand coded in PL/SQL. GetBenefits.com – Austin, TX Title CIO Time Frame April 2000 to February 2001 Designed the business process, database and the web site functionality for automated on-line teacher paycheck benefits enrollment. I directed a team of 6 programmers in Novistad Russia (Siberia) to implement the design and worked with a physical DBA in Richardson to manage the Database while I lived in Austin, Texas. We were in SQL Server. The system was OLTP rather than data warehouse. There was no ETL but there were other kinds of database centric services that were programs largely as views. Dell Financial Services Austin, TX Title PL/SQL Programmer Time FrameMarch 2000 to October 2000 I worked on a significant upgrade to the return management system. I was hired as a programmer but I was quickly informally promoted to designer. I designed the replacement data model for the return management system, implemented it and wrote the data conversion programs to move to the new design. I also helped the rest of the team of 8 re-write the SQL in the existing system and performance tune any trouble SQL statements. Dell was in Oracle and all processing was done in PL/SQL stored procedures. In fact the web user interface was generated on the fly by PL/SQL stored procedures. Americo Austin, TX Title MIS Manager Time FrameJune- to March 2000 200 employee insurance company specializing in school teacher paycheck deducted benefits. I managed a staff of 10 programmers and designed and architected a solution for the laptop sales of insurance products. I also designed, architected and project managed the development of a unified OLTP system. Our field laptop based sales system (order taker) was written with a Sybase database back end which was integrated into our back end processing systems in Oracle. We also did some data warehousing in Oracle to optimize operations. All this was of my own design since I was rather a working manager: heading IT and doing the highest level work. References PersonalDebbie Hanus- ProfessionalBrian Coats- Jason Lowe- Mumtaz Niazi - Stephanie Kennedy -