•
Build & Prototype AI Solutions
Develop and evaluate ML models, LLM-powered applications, or agentic AI workflows for a scoped business problem
Document methodology, assumptions, and trade-offs while completing a scoped analysis or prototype, then clearly translate results into actionable recommendations.
•
Develop Reproducible AI/Data Workflows
Build and document reproducible datasets, training pipelines, or evaluation harnesses
Define and document cleaning, feature engineering, prompt/evaluation, and validation processes to ensure scalability and reliability, while building reproducible datasets or pipelines.
•
Package & Communicate Results
Deliver a final package (summary, visuals, code/notebook, model card, runbook)
Present key insights, limitations, risks, and recommended next steps to both technical and business stakeholders, and deliver a final package including summaries, visuals, and supporting code or documentation.
•
Collaborate across the organization — partnering with data scientists, engineers, product managers, and business stakeholders to scope problems and ship solutions.
•
Participation in department meetings and meetings specific to the project assignment.
•
Adhering to Mosaic’s mission, guiding principles and priorities, and key competencies.