● HVAC system fault detection algorithm development based on 1-second sensor data from 3 HVAC systems in the lab (heat pump, conventional, gas furnace) and 10 residential homes. Sensors deployed included energy monitoring devices, temperature, humidity, anemometer, vibration, motion, thermostat follower, and photodetector.● Water disaggregation on 1-second interval time-series whole home water usage data(over 1 billion records) from over 30 residential homes using various algorithmic andmachine learning unsupervised techniques.● Analytics, cost modeling, and manufacturing process analysis of entire product line for a beauty & personal care products company● Analytics and cost modeling for engineering firm

● db schema development for new version (v4) of the system integrating two previous version (v2 & v3)● Team responsible for Data migration - Azure NOSQL Cassandra Application database & Microsoft SQL Server data warehouse/ analytics -> AWS PostgreSQL RDS● Write PostgreSQL stored procedures and functions to handle a significant portion of business logic and backend functionality.● Production database performance optimization using index checking and explain analyze as well as monitoring with pganalyze.

● Led thin-film photovoltaic small cell and integrated module efficiency improvement programs. Fast-paced turnaround of a program that had stalled for 18 months to near World Record efficiency in 13 months.● Directed thin film characterization metrology lab ensuring good laboratory practices and compliance. Metrology equipment included XRD, GD-OES, GTK, SIMS, FIB, and SEM.

• Manage Dataport, the world’s largest database of residential energy, water, and gas data with over 15 TB of time-series data, used by 100s of universities and companies around the globe. Over 1 billion records per month.• Solely responsible for a 10 month study of the residential energy-water nexus funded by the Environmental Defense Fund. ML cluster algorithms applied to a large (130 million records) unsupervised dataset to disaggregate water usage. A custom library of R functions created to identify and summarize energy and water events as well as calculate statistics relevant to the study. • Featured Visualizations (Google Charts, High Charts) on Dataport of large time-series energy and water dataset.• Experience writing ETL scripts (Python) to extract and load real-time whole home appliance-level energy data from IoT monitors.• Built animated admin tool (Python) to monitor and conduct remote diagnostic exploration or residential PV systems.• Hands-on experience with entire data pipeline from endpoint collection, Extract Transform Load raw data (ETL), curation and hosting (PostgreSQL, Dataport), exploration and analysis (R, Python, SQL), to visualization and knowledge (HighCharts, R, Google Charts, Python).• Offer 3rd party validation of energy disaggregation algorithms for leading energy disaggregation companies by providing and curating large-scale training and testing datasets and statistically evaluating the results.• Linux and bash scripting

• Program Manager: Led engineering and technology process improvement program through experimental planning and execution to drive baseline efficiency roadmap and process time goals on key processes supporting HelioVolt’s commercial readiness objectives (Sept 2011 - Feb 2014). • Engineering and Manufacturing interface during technical due diligence activities resulting in $50M strategic investment in Sept 2011.

• Program Manager: Led Technology and Engineering teams to develop first commercial thin-film solar module, UL and IEC certified in November 2011.• Program Manager: Responsible for the installation and qualification of HelioVolt’s Austin factory line. This program consisted of three programs; the first led by Manufacturing Engineering was the installation, characterization, and qualification of all 37 pieces of equipment (including but not limited to metal sputtering (PVD), chemical bath deposition (CBD), laser & mechanical scribing, Co-Evaporation, reaction & annealing), the second was the facilitization of all equipment led by Facilities, and the third, led by IT, was the installation and roll out the manufacturing execution system (MES). • Process/Systems: Worked with the Director of Manufacturing Engineering to develop the stage-gate tier process used to install, facilitize, characterize, and qualify all 37 pieces of equipment at HelioVolt’s Austin factory.• Process/Systems: Initiated and facilitated a cross-functional team from Manufacturing Engineering, Equipment Engineering, Supply Chain, Facilities, Process Engineering, Operations, Quality, Business Development and IT, drawing from experience and best-in-class processes to create HelioVolt’s capital equipment acquisition stage-gate process. The process has been used since 2010 to bring capital equipment projects in on time and budget.• Program Management: Brought in by the CTO to lead a cross-functional R&D team of engineers, scientists, and technicians which in six weeks developed and completed the 1364x scale-up of a new process from cell level (66 mm2) to 300mm x 300mm module on time and exceeding the efficiency target by 22%. This was the first time the R&D team hit a milestone on time.• Process/Systems: Provided PMO support for the development of HelioVolt’s NPI (New Product Introduction) stage-gate process developed by a cross-functional team of leaders from across the company.

• Solely accountable and responsible for the configuration, implementation, launch, and operation of the back-office business logic and commissions engine supporting an online dating and online travel website from Nov of 2005 to April of 2007. Responsible for scope and strategy management with stakeholders, business and functional requirements gathering and documentation, implementation and testing during transition to "go live", and the operational monitoring, validation, and reporting of key metrics.• Captured business requirements and defined metrics to create canned and ad hoc reports for the team and executive management.• Data mining and analytics to support marketing and sales campaigns• Led quality assurance and performance testing project responsible for the acceptance of a proprietary back-office business logic and sales commission engine developed in two phases by a third-party vendor and internal cross-functional team.• Led the design, implementation, and launch of an online application designed to take input from an individual’s credit report and user interface to minimize the payoff schedule for all of that individual’s debt. Due to resource constraints I was asked two weeks before launch to learn EmbPerl and implement the app, it was delivered on time.

• SA responsibilities included desktop support for company relating to hardware and software, computer applications, LAN components and peripherals, backup operations, daily system monitoring, reviewing system and application logs, verifying the completion of scheduled jobs, user account management. Supervised complaint ticketing system and ensured the timely resolution of all work orders. Provided Tier III/other support per request from various constituencies. Investigated and troubleshot issues.• Managed corporate anti-virus service solution from strategy through operation phases. • Configured and deployed internal OpenBSD firewall as part of DMZ network segment.• Part of go live team responsible for the launch of creditminders.com in 2001. On rotation for 24/7 on call support of monitoring.• Built and set-up all developer sandboxes. LAMP stack consisting of RedHat, Apache 1.3 web server, MySQL, PostgreSQL, DB2 managed database (DB2 selected), and finally EmbPerl scripting language.

• Computerized and streamlined all accounting, estimating, and project management functions into one integrated system in support of 4x growth while decreasing estimation completion time by over 50% while lowering cost variances by 56% and reducing average project completion times by two months as controller and general manager of Vernon Baker Homes.