OCI Data Science Pro case studies
OCI Data Science Pro · Put your knowledge to work in context.
The Oracle Cloud Infrastructure 2024 Data Science Professional credential, exam code 1Z0-117, is Oracle's professional-level certification for practitioners who build and operate machine learning workloads on OCI. It targets data scientists, ML engineers, and platform engineers who already work with OCI Data Science and want formal recognition of their ability to design projects, run training at scale, and keep models healthy in production. The OCI Data Science Pro case studies are 5 cases and about 4,232 words.
1Compartment, Policy, and Identity Design for a Shared Data Science Platform
Acme Retail runs a single OCI tenancy shared by three lines of business (Grocery, Apparel, Digital) and one central data platform team. Fourteen data scientists create notebook sessions and jobs.
852 words, about 4 min2Notebook Environment Drift and Reproducible Jobs for a Five-Person Team
A fraud-analytics team of five runs everything inside OCI Data Science notebook sessions on a VM.Standard.E4.Flex shape (8 OCPU, 64 GB). Over six weeks, scientists ran pip install --user to add xgboost, lightgbm, and a newer numpy; two notebooks now fail on import with a scikit-learn binary incompatibility, and one model trained in a session scores 0.79 AUC in the notebook but 0.74 from the job built from the same script.
884 words, about 4 min3Training Job Dies Mid-Run and Leaves No Artifacts
A vision team trains a 12M-parameter image classifier on 380,000 JPEGs (about 210 GB) in Object Storage. The job is configured on a VM.GPU.A10.2 shape with 128 GB block storage, and the script downloads the full manifest and decodes every image into a pandas DataFrame before the first epoch.
869 words, about 4 min4Endpoint Latency SLO Against Deployment Sizing and Bandwidth
A contract-analysis service calls an OCI Data Science model deployment over HTTPS. Requests are usually 1.5 KB, but 3% are full contracts of 6–9 MB sent as a single JSON payload.
802 words, about 4 min5Nightly Pipeline Works When Run by Hand, Fails at 02:00
A credit-risk team has a four-step OCI Data Science pipeline: read transactions from Object Storage, featurize, train, and register the model in Model Catalog, with a conditional step that deploys the new model if validation AUC is at least 0.71. The pipeline was built by a contractor in a notebook session and scheduled to run nightly at 02:00.
825 words, about 4 min