AI-12 · Advanced · ENGLISH · PLANNED
Begin with a baseline, trace a model, measure errors and improve it.
Study time is an initial activity estimate, excluding extra reading, revision and extended laboratory or project work. Videos and question bodies have not been produced by adding this outline.
Complete these courses or demonstrate equivalent knowledge. These are recommended prerequisites, not enforced enrolment locks.
YouTube playlist: ARK Artificial Intelligence | AI Engineering, Deployment and Efficiency
Playlist title reserved in the plan; no YouTube playlist or upload has been created yet.
Sign in to save this learning pathPractical outcome: Trace a model from dataset version to a monitored prediction.
Explain data versioning using a visual and one worked real-world example. Context: Trace a model from dataset version to a monitored prediction. End with a prediction or check-for-understanding.
Explain feature pipelines using a visual and one worked real-world example. Context: Trace a model from dataset version to a monitored prediction. End with a prediction or check-for-understanding.
Explain training pipelines using a visual and one worked real-world example. Context: Trace a model from dataset version to a monitored prediction. End with a prediction or check-for-understanding.
Explain experiment tracking using a visual and one worked real-world example. Context: Trace a model from dataset version to a monitored prediction. End with a prediction or check-for-understanding.
Use the first concepts from this section. Trace a model from dataset version to a monitored prediction. Produce a labelled diagram, trace or calculation; explain each step and check one changed input.
3–6 questions on the preceding first-half concepts only: recall, visual interpretation, application and one plausible misconception. Explain each answer after submission.
Explain model registries using a visual and one worked real-world example. Context: Trace a model from dataset version to a monitored prediction. End with a prediction or check-for-understanding.
Explain batch inference using a visual and one worked real-world example. Context: Trace a model from dataset version to a monitored prediction. End with a prediction or check-for-understanding.
Explain online inference using a visual and one worked real-world example. Context: Trace a model from dataset version to a monitored prediction. End with a prediction or check-for-understanding.
Explain model rollback using a visual and one worked real-world example. Context: Trace a model from dataset version to a monitored prediction. End with a prediction or check-for-understanding.
Trace a model from dataset version to a monitored prediction. Work without the example, compare two cases, record assumptions and limitations, then self-review correctness, reasoning and clarity.
6–12 questions covering Data versioning, Feature pipelines, Training pipelines and the remaining section concepts, with cumulative retrieval from earlier sections. Include a new transfer problem; 70% unlocks progression.
Practical outcome: Compare quality latency memory and USD cost for a bounded task.
Explain batching using a visual and one worked real-world example. Context: Compare quality latency memory and USD cost for a bounded task. End with a prediction or check-for-understanding.
Explain profiling using a visual and one worked real-world example. Context: Compare quality latency memory and USD cost for a bounded task. End with a prediction or check-for-understanding.
Explain gpu memory using a visual and one worked real-world example. Context: Compare quality latency memory and USD cost for a bounded task. End with a prediction or check-for-understanding.
Explain mixed precision using a visual and one worked real-world example. Context: Compare quality latency memory and USD cost for a bounded task. End with a prediction or check-for-understanding.
Use the first concepts from this section. Compare quality latency memory and USD cost for a bounded task. Produce a labelled diagram, trace or calculation; explain each step and check one changed input.
3–6 questions on the preceding first-half concepts only: recall, visual interpretation, application and one plausible misconception. Explain each answer after submission.
Explain quantised inference using a visual and one worked real-world example. Context: Compare quality latency memory and USD cost for a bounded task. End with a prediction or check-for-understanding.
Explain model compression using a visual and one worked real-world example. Context: Compare quality latency memory and USD cost for a bounded task. End with a prediction or check-for-understanding.
Explain distributed training concepts using a visual and one worked real-world example. Context: Compare quality latency memory and USD cost for a bounded task. End with a prediction or check-for-understanding.
Explain inference cost estimation using a visual and one worked real-world example. Context: Compare quality latency memory and USD cost for a bounded task. End with a prediction or check-for-understanding.
Compare quality latency memory and USD cost for a bounded task. Work without the example, compare two cases, record assumptions and limitations, then self-review correctness, reasoning and clarity.
6–12 questions covering Batching, Profiling, GPU memory and the remaining section concepts, with cumulative retrieval from earlier sections. Include a new transfer problem; 70% unlocks progression.
Practical outcome: Design a monitoring plan for drift and failed predictions.
Explain input validation using a visual and one worked real-world example. Context: Design a monitoring plan for drift and failed predictions. End with a prediction or check-for-understanding.
Explain data drift using a visual and one worked real-world example. Context: Design a monitoring plan for drift and failed predictions. End with a prediction or check-for-understanding.
Explain performance drift using a visual and one worked real-world example. Context: Design a monitoring plan for drift and failed predictions. End with a prediction or check-for-understanding.
Explain canary releases using a visual and one worked real-world example. Context: Design a monitoring plan for drift and failed predictions. End with a prediction or check-for-understanding.
Use the first concepts from this section. Design a monitoring plan for drift and failed predictions. Produce a labelled diagram, trace or calculation; explain each step and check one changed input.
3–6 questions on the preceding first-half concepts only: recall, visual interpretation, application and one plausible misconception. Explain each answer after submission.
Explain shadow evaluation using a visual and one worked real-world example. Context: Design a monitoring plan for drift and failed predictions. End with a prediction or check-for-understanding.
Explain feedback loops using a visual and one worked real-world example. Context: Design a monitoring plan for drift and failed predictions. End with a prediction or check-for-understanding.
Explain privacy controls using a visual and one worked real-world example. Context: Design a monitoring plan for drift and failed predictions. End with a prediction or check-for-understanding.
Explain incident response using a visual and one worked real-world example. Context: Design a monitoring plan for drift and failed predictions. End with a prediction or check-for-understanding.
Explain retraining decisions using a visual and one worked real-world example. Context: Design a monitoring plan for drift and failed predictions. End with a prediction or check-for-understanding.
Design a monitoring plan for drift and failed predictions. Work without the example, compare two cases, record assumptions and limitations, then self-review correctness, reasoning and clarity.
6–12 questions covering Input validation, Data drift, Performance drift and the remaining section concepts, with cumulative retrieval from earlier sections. Include a new transfer problem; 70% unlocks progression.
Combine all three sections: Design a monitoring plan for drift and failed predictions. Submit an original worked artifact, evidence or tests, assumptions and a reflection. Rubric: correctness, method, practical usefulness, reproducibility and communication. Self-reviewed, not certified.
6–12 balanced questions sampling every section of AI Engineering, Deployment and Efficiency, including interpretation, calculation or reasoning, a misconception and a new application. Passing is 70%; review errors before retrying.
Data experiments, evaluations, model tracing and reproducible projects. Guided work leads to an independent attempt, with a diagram, explanation, trace, calculation or tested artifact. Question bodies are generated when opened in your private learning path. Programming submissions are stored as text; the platform does not execute submitted code.
Use original or appropriately licensed datasets. Practical science activities use models or supplied data unless a qualified supervisor has reviewed a physical experiment.
Open the interactive practice sampler →