Hands-on service delivery in short-term contracts as an employee of a small consulting company (Infinitive) to their clients including:- top 5 bank- large pharmaceutical company ($6.37 Billion Market Capitalization)- telecommunications and media company ($54 Billion in Revenue in 2022)- medical startup (acquired by a largest diversified information, services/media company)- education startup- experience with cloud migration (AWS) and cluster- machine learning (ML) and AI space- experience with SQL- experiencw with AWS Ecosystem- advanced AI/NLP/ChatGPT- industry researchMy areas of subject-matter expertise includes skills foundational to analytics: - hands-on analysis and modeling (statistical and machine learning) - big data (Spark on Hadoop), data lakes (batch processing), stream data processing (Spark Streaming/Kafka) and enterprise data warehousing at scale (Delta tables, PostgreSQL, and Snowflake) - hands-on software development - enterprise best practices in software engineering, artificial intelligence and big data

Responsible for all aspects of "Clearcalc"(the model that extends automated loan offers to potential clients). The technology stack is Kubernetes, Docker, Python, PostgreSQL and RabbitMQ hosted in Google Cloud. My daily work includes: - machine learning (ML) in the Artificial Intelligence (AI) space - experience in deployment, cloud migration (AWS) and cluster - knowledge with Big Data/Hadoop - refactoring code to improve system design - adding new features and functionality - bringing larger parts of the system under test - migrating code from Python 3.5+ to Python 3.9 - automating a comprehensive testing suite to speed feature enhancements - mentor fellow engineers in "find the seams in the system" to work effectively with legacy code (as described in Michael C. Feathers book from 2004)

I wrote the proposal and executed the contract to add and tune Machine Learning capabilities to the Elastic Security cluster. The purpose of this engagement was to integrate advanced technology into the Cybersecurity practice to increase effectiveness and reduce the need for additional staff.Hands on delivery included:- Implemented DevOps best practices with Ansible, automated data feeds, runbook support (with standard operating procedures) and KQL training (Kibana Query Language)- Implemented a machine-learning powered enterprise SIEM solution (Security Information Event Manager) on the Elastic Stack- Configured Winlogbeat and Filebeat data ingest agents- Delivered a strategy for deploying agents customized to Jackson Family Enterprises’ needs- Configured and tuned threat detection rules and alerts- Delivered production dashboards to make threat hunting easier and enable real time data-driven decision-making- Tuned the Elastic Machine Learning configuration

Leveraged over a decade of experience in data architecture, data modeling, and enterprise data lake construction to store and process raw multi time zone data and made it ready for analysis and modeling workloads such as: Data Warehouse cross-tabulation reporting; Power BI visualization and dashboarding; Machine Learning and Statistical Modeling; Elasticsearch search engine ingest for discovery/dashboarding, and Data Warehousing in PostgreSQL and Parquet/Spark/Hadoop for analysis and reporting. Personally delivered all aspects of the people (mentoring/training) process and tools strategy to open new lines of business for the organization, including training director-level staff on project portfolio management. Delivered a cloud-first data engineering solution that takes raw on-prem data and ingests it in analysis-ready form into the AWS data lake for Data Warehouse analysts and Data Scientist Machine learning efforts. Consult with C-Suite executive leadership to build business capturing capabilities with state-of-the-art Big Data Processing (Hadoop 3/Spark 3/Scala 2.12) capabilities and building our bench of future machine learning professionals leveraging both Scikit Learn and AWS SageMaker. A typical day for me is spent briefly on DevOps tasks before grabbing the next dataset and following my documented process of exploratory data analysis, feature engineering, and machine learning modeling to deliver ROI on company data. I use these processes to mentor and grow our bench as I believe machine learning should be as accessible and as available as a pivot table in Excel to our team.

As the company's only Data Scientist, I built all of their ML and Big Data capabilities from the ground up, including their AWS cloud implementation. Defined organizational roles and responsibilities, designed/created data product deliverables, created all business processes, conducted staff seminar, mentoring, training and advised C-Suite executive leadership on how to open new lines of business. Introduced the concepts of information technology portfolio management, detailed project planning and earned-value execution with Agile methodology, and personally managed data engineering infrastructure and system delivery. Within the first year I saved a $50 million customer contract in three weeks for approximately $2,500 in cloud fees. Designed, implemented and supported ETL data pipelines in Spark using Python/Scala as well as the down-stream Spark Analytics (also written in Python/Scala). Further analysis and reporting was done in R with ggplot2 and would later be migrated to Python/Pandas/Matplotlib. Implemented AWS VPC, EC2, SSG, IAM, S3, and EBS. Integrated AWS services with Python and boto/boto3.Provided professional services from my own company (C9 Tech, LLC) to build ML and Big Data capabilities. Built a basic data warehouse with AWS S3, Parquet, and PySparkSaved a $50m company contract by delivering a 100% custom in-house data analysis solution within three weeks. Developed ETL data pipelines in Spark, advanced analytics, and machine learning in Spark using Python and Scala Visualization and data analysis in R/ggplot2 (primary) and later migrating to Python/Pandas/Matplotlib. Began the process of building a data lake and demonstrated ROI through data feeds to different systems. Trained engineers in both data science and software engineering.

Supported large federal agency leading a team of engineers delivering custom solutions. Managed a $20 million/year software project portfolio providing subject matter expertise to managing the three subcontractors (IBM, Boeing, and Computer Science Corporation). Applied advanced statistical modeling with R doing visualizations in Gnuplot (highly underrated visualization software). The nature of operations changed with predictive models I produced to help the customer accurately predict failures and quantify the performance impact within their data center portfolio.The customer deployed me where I was needed most including two years on the Oracle DBA Team doing application development, SQL Analytics/Reporting, implementing data science solutions from scratch (like a collabortive filter/recommendation engine). The common thread among my roles was processing massive amounts of information to deliver actionable insight to decision-makers.

Supported metro DC federal agency, leading team of engineers, creating demanding Java EE web applications building a large enterprise application with Java Server Pages, Java Server Faces, XHTML/CSS/Javascript, and Java Beans using the Model-View-Controller design pattern (which was not completely embraced at the time). Wrote C/C++ Solaris/Linux system programming code to address performance sensitive requirements. Utilized the Java Native Interface (JNI) to integrate C code with Java.Optimized configuration management to govern our Subversion repositories, organizing code into branches, and setting branch policy. Over a weekend I piloted some (early) CI/CD solutions before selecting a relatively new product called Hudson (now called Jenkins). I set up my code with an automated unit test suite (JUnit, CPPUnit) and learned how to run a suite of code quality checks (e.g. unit tests, linting, code coverage with Cobertura) and provide feedback on build-breaking code check-ins within two minutes.All of this took place around the time the Defense Industry was applying the Capability Maturity Model (CMM/CCMI) to Agile Software Development projects and aggressive automation via CI/CD pipelines were the only way that was going to work within a single sprint.

Supported large metro DC federal agency working on a C/Python distributed application that strengthened my ability to rapidly identify Linux and custom software bottlenecks using tools like iostat, vmstat, ltrace, strace, netstat.I transferred to a better fit within Northrop Grumman where I worked on Java EE application programing using the vaunted Java Server Pages (JSP) to build enterprise web applications with this new technology (to us) called AJAX to enhance our HTML/CSS/Javascript application -- pretty exciting stuff for the enterprise in 2006/2007. Model-driven development was all the rage using the unified modeling language (UML) back then. Worked on projects with Oracle Applications r11 (now called Oracle Fusion).

Supported intelligence community with custom software engineering solutions. I will always be grateful to Lockheed for showing me a whole world of enterprise technology including IBM Z-Zeries Mainframe running MVS, Safety-Critical launch vehicle tracking software for Vandenburg AFB and Cape Canaveral, C programming at scale I had never heard of, and the concept of having a meantime between failure of 25 years. Lockheed introduced me to the mindset of knowing your platform (Unix/Linux/MVS) extremely well and "extending" functionality through system programming to create an integrated system whose value as a whole was greater than any one part. I became a technical leader and learned to love/hate C++ programming on AIX/Power 3 using Visual Age for C++. I also learned ClearCase with a healthy fear of the "evil twin" problem.I had the good fortune to work with former IBM Federal employees who showed me how to write production-grade Unix Shell Scripts and how they organize modules in a large-scale C/C++ system. Lockheed introduced me to the mindset of knowing your platform (Unix/Linux/MVS) extremely well and "extending" functionality through system programming to create an integrated system whose value as a whole was greater than any one part. I studied and worked hard to become the technical leader Lockeed deserved for giving me an opportunity. I learned to love/hate C++ programming on AIX/Power 3 using Visual Age for C++. I also learned ClearCase with a healthy fear of the "evil twin" problem.I had the good fortune to work with former IBM Federal employees who showed me how to write production-grade Unix Shell Scripts and how they organize modules in a large-scale C/C++ system.

I supported the finance industry writing custom software and spent my days living in Oracle SQL*Plus in a Solaris terminal learning just how different Oracle 7 is from Microsoft Access. I was lucky to have some industry veterans explain why PL/SQL was a thing and amaze me with the kind of data they ingested. Supporting an end-of-lifed version of Oracle was impressive but supporting reel to reel tape ingest is amazing. To this day nobody has shown me a stronger grasp on reel to reel tape and "block density" than Mike Fusa. I was awe of the SUN e-450 everything ran on in its two processor and 4GB of RAM configuration. Rhonda Mattana showed me how she wrote SQL reports to advise the C-Suite. It was this experience that taught me how critical trust and relationship building with the C-Suite is to advise decision makers and provide a return on investment from a technology portfolio. Rhonda showed me that actionable insight is the real end-goal from analysis.

While in college, provided consulting services with programing and analytics to PRG/Shultz, a firm auditing Staples. I provided on-site SQL analytics and mined their data looking for payment anomalies.