Oracle Cloud Infrastructure Data Science Professional (2025): Hands-on Workshop

Course Description

Oracle Cloud Infrastructure Data Science Professional (2025): Hands-on Workshop focuses on how teams design and operate OCI Data Science projects, notebook sessions, conda environments, Accelerated Data Science SDK, Data profiling, preparation, feature engineering. Learners connect configuration choices to user responsibilities, data behaviour, security boundaries, and the business result that each workflow is expected to produce.

Practical work extends to Model catalog, metadata, artifacts, model deployments, Jobs, pipelines, schedules. Evidence such as working notebook environment and reproducibility record, feature pipeline and experiment comparison report, registered model and endpoint validation trace, automated ML workflow and operational monitoring plan is used to verify results, test edge cases, explain exceptions, and support a controlled move from classroom scenario to production practice.

After the four lessons, learners can help establish a reproducible OCI Data Science workflow that prepares data, trains and deploys models, automates lifecycle tasks, and monitors business-facing predictions. They also document the next improvements needed to strengthen control, usability, resilience, data quality, and support.

What You Will Learn

• Explain how OCI Data Science projects, notebook sessions, and conda environments establish the operating foundation and shape later configuration decisions.
• Configure Accelerated Data Science SDK, object storage access, repositories, and IAM, then confirm readiness through working notebook environment and reproducibility record rather than relying on screen messages alone.
• Execute a controlled workflow covering Data profiling, preparation, and feature engineering, preserving input traceability, approvals, and the resulting changes.
• Read feature pipeline and experiment comparison report to isolate cross-validation results overstating performance because data leakage was introduced during preparation, separate symptoms from causes, and select a correction that can be safely retested.
• Integrate Model catalog, metadata, and artifacts with ownership, access, and downstream process expectations.
• Challenge model deployments, endpoints, scaling, authentication, testing, and prediction integration using normal and boundary cases, then defend the outcome with registered model and endpoint validation trace.
• Investigate a scheduled retraining pipeline publishing a weaker model without a promotion gate by reviewing Jobs, pipelines, and schedules before applying a reversible fix.
• Prepare automated ML workflow and operational monitoring plan that demonstrates MLOps automation, drift monitoring, logging, cost management, troubleshooting, and professional exam scenarios and supports handover, monitoring, and the next improvement decision.

SkillDynamo Training Approach

Training uses instructor walkthroughs, hands-on configuration, transaction-based exercises, boundary testing, troubleshooting drills, validation checklists, and structured handover. Learners practise explaining why a result is acceptable, not merely showing that a screen was completed.