Predictive Analytics, Confidential Data Confidential (Postgres Core), Google BigQuery, Linux, PERL, KXEN, MySQL, Vertica, InfoBright, Kettle (ETL), Aqua Data Studio, Confidential, Python, Ruby (Rails), JQuery, Sybase, SQL Server, NoSQL ( Confidential, Hadoop, MongoDB, Confidential, FastBit), PHP, Oracle, PL/SQL, Cloud computing, Pentaho.
Solution Architect/Consultant, San Francisco, CA
- Design multiple OLAP data models for consolidation and aggregation of financial data currently residing in multiple OLTP MySQL database tables. Guide the financial analysts on best practices for pulling/migrating/combining data from one system to another.
- Create complex data structures for multiple Executive Dashboard reporting presentations.
- Wrote 30+ bash shell scripts to automatically pull and load (ETL) the data via cron under Cygwin on a Windows 7 system and finally migrating to a Linux server to handle the cron tasks and development environment.
- Became the de - facto DBA for the Google Cloud MySQL installation as well as data processing on Google’s BigQuery Analytics Platform. Migrated all local data stores/structures to the cloud platform.
- Developed a complex auditing process on game activity required by internal and external audit teams.
- Document implemented processes/methodologies and best practices.
- Served as a technical mentor for the finance team.
- Installed 4 Confidential clusters across CentOS and RHEL Linux environments for data analytics.
- Data migration from SQL Server DB to Confidential to take advantage of Confidential 's OLAP methodology/clustering platform with the data modeling in AquaDataStudio.
- Large-scale migration project to roll into production a successful Confidential POC.
- 12 heterogeneous bank systems of source were combined into one homogeneous data structure in an Confidential database cluster that will eventually contain ~25 months of data for extensive data analysis.
- The data is sessionalized and made available for various Confidential analytical activities.
- Wrote ~25 bash ETL/ELT shell scripts for data migration from the source systems. Data models built using both Confidential and AquaDataStudio.
- Additional data sources will also migrated over the next year.
Confidential, San Francisco, CA
- At Confidential I helped build/deliver a reporting demonstration for Teradata/Aprimo via Tableau sitting on a Teradata- Confidential database cluster containing ~1 billion data rows. This included ETL processing as well as building the analytical processing to support a dozen various Tableau reports.
- Using Confidential 's built-in analytical functions, nPath and Attribution, the reports demonstrate marketing statistics/analytics that assist in assessing various business marketing/systems targeting efforts.
Confidential, San Francisco, CA
PHP/ Confidential /MySQL Consultant
- Assumed responsibility for a complex PHP-base demographics based advertising targeting process.
- Using numerous PHP scripts that access US and State Census data the process derives tracts/boundaries that comprise population demographics across six categories (age, gender, income, education, marital status and ethnicity).
- Additional PHP processes calculated each tract’s tally for each of the various demographic categories and loads them into a Confidential nosql data cache.
- The cache is then uses to calculate the percentage of each category for a given venue.
- A venue may cross multiple tracts/boundaries so the calculations can be complex.
- The resulting statistics are used to determine what the audience would be for a given venue in a geographical location.
- This helps determine which ad to place at that venue which are actually video monitors that display video advertising.
- The results are loaded into another datastore, FastBit, for access by the network processes.
- I rewrote/modified the calculation processes to be multi-threaded as this process could take days to complete.
- It now just takes hours to process a full set of venues. Currently working on an API that accesses the calculation process via MySQL triggers that queue up venues that are new or modified and need to re-calculate.
Confidential, San Francisco, CA
CTO/Architect/Principal Software Engineer
- For this website, I directed the technology for and built the website from the gathering of 25k+ food products, ETL processing and data modeling via Confidential, and web interface with the artistic and informational input from four other collaborators/partners.
- Created and launched the first and only nutritional website that compares additives, ingredients and nutritional facts in packaged foods. On the site, product ingredients and nutritional facts are linked to proprietary glossaries in order to help users decipher food labels.
- Created a datastore in Confidential that tracks advertising assets and their performance. The initial purpose was to leverage a NOSQL datastore for advertising optimization.
- Developed using Ruby and a third-party portal to retrieve data.
- Once the data was in Confidential the next step was to devise logistical regression processes to monitor and enhance asset performance.
- The project was put on hold until an enterprise-wide Confidential implementation could be installed.
Consultant - Data Warehouse ETL Architect/Developer
- Developed large scale ETL/ELT processes for migration project from MS SQL Server on Windows platform to MySQL database on LAMP in Amazon Cloud with Kettle ETL toolbox. ERD and data modeling via Aqua Data Studio. DBA for the cloud based MySQL servers with replication. Project ran out of money.
Confidential, Santa Clara, CA
Data Warehouse Architect
- Migrated existing Customer, Order, Financial MySQL systems to Confidential distributed MPP database system.
- Designed, modeled ( Confidential and ADS) and built the Confidential database processing and platform in the Amazon Cloud.
- Migrated all the legacy ETL bash shell processing to PERL incorporating the built-in Confidential loading tools.
- Created a number of data validation and correcting utilities to identify and repair corrupted data/structures in the legacy systems.
- Assisted in the hiring and training of new staff to build out the new Confidential based database system and the creation of processes utilizing the Confidential built-in regression/analytics toolset including Confidential ’s nPath time series analysis and Map-Reduce/Hadoop processing.