iOS AI/ML

Acknowledgement: Shared by Peter Bamuhigire, techguypeter.com, +256 784 464178.

Use When

• Building or reviewing iOS, iPadOS, or macOS AI/ML features using Foundation Models, Core AI, Core ML, Vision, NaturalLanguage, Speech, SoundAnalysis, Create ML, Evaluations, Private Cloud Compute, or a third-party Language Model provider.
• The work includes on-device inference, custom models, multimodal prompts, agentic model tool use, AI evaluation, model-provider selection, privacy-sensitive AI, or Apple Intelligence availability gates.

Do Not Use When

• The task is plain iOS implementation with no AI/ML path; use ios-development.
• The task is AI product architecture outside Apple platforms; use ai-app-architecture, ai-llm-integration, or ai-evaluation.
• The task is only AI security, prompt injection, or action authorization; use this skill with ios-security-and-rbac.

Required Inputs

• Feature goal, model/provider choices, target Apple platforms, device/OS floor, privacy constraints, offline expectations, data inputs, tools/actions the model may call, and evaluation criteria.
• Confirm whether the deliverable is design, implementation, migration, review, evaluation plan, or release evidence.