opensite-ai/opensite-skills / aws-ai-ml
aws-ai-ml
VerifiedSelects, deploys, and customizes AI models on Amazon SageMaker. Training or Processing jobs, fine-tuning (SFT/DPO/RLVR/RLAIF), model selection, dataset preparation, evaluation, SageMaker or Bedrock deployment, inference optimization and endpoint diagnostics. Covers the full lifecycle from planning through production. Use when fine-tuning models on SageMaker, choosing which base model to customize, fine-tune, or deploy from SageMaker JumpStart or Hub, SageMakerPublicHub, or the SageMaker public model catalog, transforming or validating training data, evaluating model quality, deploying or optimizing endpoints, configuring IAM/S3 for training, or managing SageMaker Managed MLflow. Use for endpoint health, failures, latency, logs, metrics, errors. Covers Serverless Model Customization, Nova and OSS deployment paths, and PySDK v3. NOT for Ground Truth labeling, Feature Store, or general-purpose AWS infrastructure.
references/use-case-specification/references/spec-for-finetuning.md
View on GitHub# Spec for Fine-tuning
## Phase 1: Discovery (1–3 turns)
Review what is already known from the conversation so far, then identify what is still missing. You need these three things:
- **What** is the problem the user is trying to solve with model customization
- **Who** will use the finetuned model and in what context
- **Which** success criteria can be used to evaluate how well the custom model performs compared to the base model on a test set. Success criteria must be measurable by an LLM-as-a-Judge (e.g., response accuracy, tone adherence) — not things like latency or throughput.
**Guidelines**:
- Infer as much as possible from what the user has already said
- If the user gave examples, use them to fill gaps rather than asking again
- Only ask clarifying questions when you cannot infer the information needed for Phase 2
- If everything is already clear, say "You've given me a clear picture. I'll put together a use case specification now." and move to Phase 2.
⏸ Wait for user after each clarifying question.
## Phase 2: Produce Spec
1. Save all generated artifacts under the project directory structure defined by the directory-management reference, if available.
2. Synthesize the information into a Markdown document called `[relevant_title]_use_case_spec.md`:
```markdown
# Use Case Specification
## Intent
Fine-tune
## Business Problem
[Concise problem statement + what the custom model will do]
## Primary Users
[Who uses the model and in what context]
## Success Tenets
1. **[Tenet Name]** — [Description of what success looks like, measurable by LLM-as-a-Judge]
2. **[Tenet Name]** — [Description]
3. **[Tenet Name]** — [Description]
```
3. Present the spec to the user:
> I have put together a use case specification and saved it in [filename].
>
> A use case specification is a design principle recommended by the [AWS Responsible AI Lens](https://docs.aws.amazon.com/wellarchitected/latest/responsible-ai-lens/design-principles.html).
>
> [use case in human-readable format]
>
> Does this match your intent?
⏸ Wait for user approval.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| writing-skillsobra/superpowersUse when creating new skills, editing existing skills, or verifying skills work before deployment | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| xlsxanthropics/skillsUse this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .xltx, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| shipping-and-launchaddyosmani/agent-skillsPrepares production launches. Use when preparing to deploy to production, or when asking what needs to be in place before shipping. Use when you need a pre-launch checklist, when setting up monitoring, when planning a staged rollout, or when you need a rollback strategy. | Stars: 119k | Updated | Claude CodeCodex | Verified | Install |
| childhoodthedotmack/claude-memBabysit a shipped release after merge, publication, or deployment. Verify published packages, production behavior, installed plugins, and late CI or user reports until the observation window is clean; use babysit for PR readiness before release. | Stars: 118k | Updated | Claude CodeCodex | Verified | Install |