Data Scientist
Summary
This position primarily involves work in the Analytics field and working with analysts, performing data definitions, data extraction, transfer and loading, data preparation, data validation, issue/problem identification and resolution, data cleansing and other processing, filter application, reference class development, testing and deployment, verification, smart report development, implementation and testing, specification development, vendor interfaces and communications, descriptive analyses, predictive modeling, project/scenario simulation, testing, and interfaces and communications with other groups including information technology, data support, information management/systems, and various project groups, locally and remotely (both in and out of country).
Scope
This role supports the development, testing, implementation, and maintenance of the Fluor Big Data Analytics Environment and System(s). The Data Scientist I role takes direction from the Lead Data Scientists and Analyst(s) as well as management in order to assist in the work to design, implement and test various types of statistical analyses and prediction models, including understanding the existing systems and the investigation, development, testing, quality inspection, and validation of new systems when that is a requirement. The Data Scientist role requires the development of a deep understanding of the data, processes and systems, to set up various systems of analysis and their smart graphics to enable knowledge, insights/indicators to be extracted as information from the data input, and presented to the users in various forms. The data may be obtained, for example, from structured and/or unstructured formats and a wide variety of sources. This position requires the candidate to work in a highly collaborative environment with other data scientists, analysts, some key vendors, and company management. Being a data scientist support the process to provide clear, easy to use and access, analytics communications that enable deep investigation by the projects (management and project controls), ultimately to drive project execution performance results for the company.
Principal Job Duties & Responsibilities
• Improve and/or develop SQL/R/python algorithms for data extraction, transformation, and cleaning of the structured and unstructured data from Fluor project execution & business tools /systems;
• Improve and/or develop and support the data preparation, R/Python/Plotly/LWRS code for smart graphics and analytics charts (for descriptive, predictive and operational requirements from mgmt and analyst teams);
• Integrate data sets; Develop or program databases and query them to retrieve data to perform statistical analysis and to create reference class cohorts of data as may be required;
• Use statistical analysis and programming to understand and work around possible data limitations
• Brainstorm with functional experts in performing exploratory data analytics to diagnose the execution issues in the large multi-billion-dollar capital projects, and develop the related visualizations required for understanding the results, using tools such as R and others and its visualization packages such as gg plot, plotly shiny apps, LWRS, etc.
• Responsible for development of analytics charts from start to end including specification package development and review/explanation, Data Preparation, visualization development using R, Java Script, Plotly, testing and review with analysts. The ability to write JSON objects to transfer data and JavaScript code to UI, and maintaining code using Git version control are valued assets.
• Develop new prediction models or enhance & maintain existing prediction models predicting the future outcomes of capital projects using machine learning techniques e.g., Random Forest, Naive Bayes, SVM, Elastic Net, and XGBoost. Test models for drift and resolve along with other issues and problems.
• Perform QA testing for the EPHD UI / prediction models / Chart development work, done by the vendor and/or as peer review to Fluor data science colleagues
• Perform web usability testing, functional testing, and performance testing on Dev, QA and Production environment.
• Work autonomously and in a highly collaborative team environment (sometimes as support and sometimes as lead as you grow into various roles and take on more responsibility over time).
• Be highly motivated and a self starter. Ask questions. Love to learn and grow and provide support as a service to teammates and projects.