Data Scientist with Geospatial

Skills

Description

Candidate would be focused on processing and analysis of geospatial on-farm data sources, using at a minimum Midas and Trino databases to extract reliable, actionable information from these data. Python and pipeline coding for Midas Engine and the use of Midas Model API would be key tools. Additional tools that may be required could include the weather pillar, GSI API and NSL resources.

Candidate would be asked to develop and implement algorithms and strategies that improve on-farm data quality, extract on-farm data value and leverage it for added value to related Corteva toolsets.

High level of skills with Python and geospatial applications. Ideal candidate would have some experience in precision agriculture or related technologies. Number of years of experience is less important than the scope of experience in this role.

Responsibilities

• Experience building models that solve business problems
• Experience independently answering business and related analytics questions with interpretable and direct quantitative approaches.
• Experience in using Python for analysis, including the use of common libraries (pandas, numpy, scipy, matplotlib, seaborn, etc.), creating and sharing jupyter notebooks, and building reusable functions and scripts.
• Fluency in using SQL for common data access, transformations, and aggregations.
• Ph.D. or M.S. in Statistics, Data Science, Computer Science, Physics, Engineering, Operations Research, or related scientific discipline.
• MUST have strong communication and ability to translate technical components into non-technical terms.
• Ability to tell a story about what's happening with a data model, why, and ability to reinforce the decision or pivot.
• This leader will look for strength + passion in problem solving, finding an answer using a toolkit/statistical tools; problem solver and guiding solutioning for next model. For example, if they find Model X is not right for a group of users, investigate another solution - where is it not perfect, be able to tell the story why; SME - Hypothesis forming.
• Does not need to have Ag experience. Always helpful, but not necessary to be considered.
• PHD is not required – MS is fine. MUST have 2+ years of work experience
• Self-starter confidence – reach out for help when needed but be able to work independently
• MUST be strong in SQL, Python, Pandas, and strong clean coding practices - checking for errors, duplicates, 1:many ratios, etc.
• Ideally they have some stats background
• NOT looking for someone who just wants to run and tune data models - this role will be full stack, finding data sources, normalizing, running analysis on the model and telling a story about it. May partner with someone else on the team too.