Bryan Curtis

What I specialize in

I specialize in water data operations across the enitre water data journey --- data loading, transformation, analytics (statistical machine learning and AI), visualizations, and end-user applications --- and the full water analytical development cycle --- innovation, development, and integation --- with DataOps principles of agility, statistical process controls, and collaboration to rapidly deliver engaging and error-free applications for water insights. 

I am particularly skilled in automated quality control and uncertainty estimation, applying my skills and insights across the entire data - development - operations stack and piplelines. I lean heavily on my scientific mindset, statistical rigor, and deep context awareness to apply these skills thoughtfully in ways that maximize application success.

Previous projects

As an independent software developer, I'm building water supply and demand forecasting machine learning models for realistic spatio-climate-economic-behavioral scenarios and lightweight/portable/reproducible residential + commercial + agricultural water customer engagement apps with SQL, R, Python and REST APIs. Impact -> clear and correct insights that all water managers in any sector and the public can use to make intelligent water decisions.

As a Water Conservation Data Engineer/Analyst, I developed water conservation analytics and customer engagement applications (e.g. water bill forecasting tool) for water utilities that was built within a data pipeline operation that I setup, reshaping a legacy water utility information system into an open, reproducible, and modern data operation. I reported water utility water loss metrics to the State of Colorado by collaborating with internal water resource and operations teams, compiling and quality checking utiltiy water production and customer water use data, and developing in-depth knowledge of utility water balance monitoring.

As a Weather Data Scientist, I created an end-to-end quality assurance and control system for a meteorological observation network of over 220 weather stations across the State of Washington. I quantified uncertainty in weather observations through direct hardware inter-comparisons and statistical models (probabilistic and machine learning) informed by climatology and terrain complexity. I maintained and updated MySQL databases to optimize data storage and engineering for server applications. I wrote reproducible, contained, and automated quality control algorithms (e.g. station specific range limits based on climatology), quality assurance checks (e.g. comparisons to gridded weather forecast model output) and data applications/visualizations (e.g. network wide station correlation tool) in R and Python that run in a cloud environment on real-time data streams.

As a Graduate Research Assistant and Master's student at the Institute of Arctic and Alpine Research, I applied advanced Bayesian machine learning methods to correct precipitation measurements in the windy sub-alpine and designed, installed, and regularly sampled a snow course network of 45 snow board sampling locations for storm specific measurements of snow water equivalent (SWE) and snow depth across the Como Creek watershed between the CU Boulder Mountain Research Station and Niwot Ridge, Colorado.

As a Senior Research Associate with the Lawrence Berkeley National Lab, I deployed field micrometeorological sensing (CO2, H2O, CH4, and radiation) platforms to measure the effects of permafrost thaw on local hydrology and the near-surface atmosphere, providing critical data for climate policy and modeling.

Throughout my career as a scientific researcher, I have contributed to multiple high-impact publications and scientific conferences on land-atmosphere climate feedbacks of carbon, water, and energy.

How can I help you

I can help build and operate lean, agile, performative, and error-free data systems for water operations at any scale for any purpose, particularly for water operations concerned with water resiliency and interest employing innovative analytics and intelligence. My extensive experience with environmental monitoring and sensing can be leveraged to assist in smart and distributed data collection networks to load strategic data into robust data operation pipelines. My analytical background can be used to build powerful machine learning models and combined with my proficiency in customer engagement applications, can help deliver accessible, intuitive, and insightful applications to stakeholders, customers, and end-users. Critically, my attention to detail and statistical process control approach can help create data operation systems that can scale, iterate, and deliver correct data-driven insights quickly and often.

I can help turn challenges into opportunites and create water positive outcomes.