Principal Applied Scientist

We are searching for a Principal Applied Scientist to join OCI Planning Strategic Initiative (PSI) team in Nashville, TN this is a critical role in improving operational efficiency, driving cross-functional visibility, and enabling leadership to make data-backed investments and risk-aware decisions. The ideal candidate will be a highly motivated, analytical individual with strong communication skills to help the organization meet its strategic and financial objectives. Responsibilities will include translating disparate and complex data sets into actionable insights across OCI’s strategic initiatives. The ideal candidate is a subject matter expert in business operations and analytics, with a passion for data and a strong ability to translate complex information into actionable insights. D eep technical leadership to forecasting, scenario modeling, optimization needed to improve planning accuracy, quantify risk, and guide high-stakes capacity and investment decision at OCI scale. This leader will set scientific direction, establish rigorous evaluation and governance, and partner across planning, engineering, and finance to operationalize decision models into repeatable planning processes and measurable outcomes.

Responsibilities:

Demand, growth, and workload forecasting (multi-level)

• Develop and improve forecasting models for OCI demand/consumption and key drivers (by region, service, customer segment, shape/SKU)
• Build hierarchical approaches (global>region>AD/cluster/service) and reconcile forecasts across levels.

• Quantify and communication uncertainty (prediction intervals, scenario ranges) to support planning decisions.

  Capacity planning and supply risk modeling

• Translate demand forecasts into capacity requirements (compute/storage/network) considering constraints (lead times, redundancy, placement rules)
• Model Supply risks and constraints (procurement delays, buildout timelines, utilization targets, failure domains)
• Create risk indicators (headroom depletion probability, hotspot detection, time-to-exhaustion

Optimization & Decision Analytics for Planning

• Formulate optimization problems to recommend actions (where to add capacity, how to allocate limited supply, prioritization across regions/services
• Perform tradeoff analysis across cost, performance, availability, strategic priorities.
• Develop scenario tools for “what if” planning (demand shocks, product launches, regional events)
• Measurement, back testing, and model governance

Data partnership & Feature/driver engineering

• Identify and validate leading indicators (pipeline signals, launches, pricing changes, customer ramp patterns, seasonality, macro factors where allowed)
• Improve data quality through anomaly detection, reconciliation, and automated checks.
• Explore and research data and answer strategic business questions, with data transformation, statistical analyses, or machine learning methods
• Fine-tune and optimize algorithms and models to ensure scalability, reliability, and performance at a high level

• Collaborate with data scientists and other stakeholders to understand data needs and develop solutions that meet those needs

  Cross-functional influence and communication

• Partner with planners, finance/FinOps, capacity engineering and product teams to align on assumptions and actions
• Communicate results to exec and operational stakeholders with clear narratives: drivers, uncertainty, risks and recommended mitigations.

• Drive adoption-ensure models change planning decisions and improve outcomes.

  Metrics and KPI design expertise , including defining the north star and guardrail metrics and ensuring metric definitions are unambiguous, consistent, and actionable.

  Basic applied ML understanding , familiarity with model types (classification/regression), evaluation metrics (AUC, precision/recall), and core concepts such as feature engineering, model training, offline/online evaluation, deployment/monitoring and iterative improvement. ML algorithms, feature engineering, model training, offline/online evaluation, deployment/monitoring and iterative model improvement

  Basic Data Engineering , including building data pipelines, reliability/SLAs for data, governance/definitions, reporting layers, and operational analytics

  Experimentation (A/B testing) experience , including designing experiments, analyzing results, avoiding common pitfalls

Qualifications:

Proficient in SQL and/or Python development for data analysis including writing complex queries, joining multiple sources, validating data quality, and building analysis-ready datasets; python for analysis using pandas, notebooks, visualizations, and basic modeling.

Experience with ETL/ELT and data modeling concepts such as dimensional modeling, event-event-log design, and semantic layers

Experience building dashboards and BI reporting (e.g., Tableau/Power BI/Looker) to operationalize insights and support decision-making.

Metrics and KPI design expertise , including defining the north star and guardrail metrics and ensuring metric definitions are unambiguous, consistent, and actionable.

Basic applied ML understanding , familiarity with model types (classification/regression), evaluation metrics (AUC, precision/recall), and core concepts such as feature engineering, model training, offline/online evaluation, deployment/monitoring and iterative improvement. ML algorithms, feature engineering, model training, offline/online evaluation, deployment/monitoring and iterative model improvement

Basic Data Engineering , including building data pipelines, reliability/SLAs for data, governance/definitions, reporting layers, and operational analytics.

Experimentation (A/B testing) experience , including designing experiments, analyzing results, avoiding common pitfalls (selection bias, peeking, multiple comparisons)

Business/product problem framing with demonstrated ability to translate ambiguous questions into measurable hypotheses and requirements, and communication tradeoffs clearly

Strong stakeholder management and executive communication skills , including synthesizing complex analyses into clear recommendations for executive audiences.