Kit · product teams
Product model selection kit
Turn a product use case into a reviewable model selection artifact. Walks the product manager learning path, four lessons covering use-case framing through benchmark limits, three exercises that produce pricing + lifecycle notes + the brief, the prompt-testing + long-context playbooks, two prompt sets, and the model evaluation plan template.
Audience
product teams
Difficulty
intermediate
Estimated time
180 min · 7 steps
Who this kit is for
Use-case framing, pricing references, lifecycle gates, and Markdown evidence briefs for product managers and technical buyers. The platform surfaces verified fields; your team owns the decision.
Goal
End with a use-case shortlist, a pricing-reference note, a lifecycle risk note, a comparison URL, and a Markdown evidence brief the team can review together.
What you will produce
- Use-case shortlist URL
- Pricing-reference note
- Lifecycle risk note
- Model comparison URL
- Decision evidence brief
Prerequisites
Export Markdown
The kit serialises to a single Markdown document you can paste into a design doc, ticket, or PR description.
Workflow
7 sequenced steps
Open each step in order. Every step opens a route that already exists — no parallel UI.
Step-by-step workflow
Walk the product manager learning path
Read the four lessons (how-to-choose-ai-model, pricing-references, model-lifecycle, benchmark-limitations).
Open
/learn/path/product-manager→Output: Notes covering use-case framing, pricing references, lifecycle, benchmark limits.
Open the governance-review shortlist
Open /select?useCase=governance-review (or a use-case relevant to the product) and save the URL.
Open
/select?useCase=governance-review→Output: Use-case shortlist URL that opens the same view for the team.
Review pricing references
Capture the candidate pricing rows the finance projection will depend on.
Open
/learn/exercises/review-pricing-reference→Output: Pricing-reference note (provider + unit + retrievedAt).
Inspect lifecycle
Capture the candidate lifecycle field, any retirement date, and the provider's named successor (if any).
Open
/learn/exercises/inspect-model-lifecycle→Output: Lifecycle risk note for the integration plan.
Run the prompt-testing playbook
Run a minimum prompt-test routine against the candidates in your own harness.
Open
/lab/prompt-testing-basics→Output: Per-prompt observations to attach to the brief.
Fill the model evaluation plan template
Adapt the model evaluation plan template; copy in the shortlist, pricing, lifecycle, and observation notes.
Open
/lab/templates/model-evaluation-plan→Output: Markdown plan paste-ready for review.
Generate the decision evidence brief
Open /briefs/build with the candidate slugs and export Markdown for the review meeting.
Output: Markdown brief with verified fields + data gaps + source trail + freshness.
Required resources
The kit reuses existing surfaces — no parallel UI, no duplicated content. Open each surface in the order the timeline lists.
Lessons
How to choose an AI model →
A workflow for picking which AI model to test next — start from your use case, inspect verified fields, export an evidence brief.
AI model pricing references explained →
Why catalogue pricing rows are references, not quotes — and how to read them without ranking models by price.
Model lifecycle: active, deprecated, retired →
What active, preview, deprecated, and retired mean for a model — and why lifecycle should gate integration decisions.
Why benchmark scores can mislead →
Contamination, prompt variance, version drift, and why the catalogue does not publish provider-reported benchmark scores casually.
Exercises
Review a pricing reference safely →
Open a verified pricing row, read its unit semantics + retrieval date, and walk the reverification queue if it is stale.
Inspect lifecycle before integration →
Pull lifecycle state for a candidate model, check for retirement date, and add a migration target to your notes if one exists.
Create a decision evidence brief →
Use the decision brief builder to generate a paste-ready evidence pack from your shortlist, then export it in Markdown.
Lab playbooks
Prompt testing basics →
The minimum prompt-testing routine to run against a shortlisted model before integration. Defines a representative prompt set, structured observations, and concrete failure modes — no benchmark scores.
Long-context testing →
How to test long-prompt behaviour past the catalogue's verified context window — recall, instruction adherence, and cost growth — without trusting a marketing number.
Evaluation prompt sets
Final checklist
- Use-case shortlist URL is saved.
- Pricing-reference note has unit + retrievedAt.
- Lifecycle risk note has the retirement date if any.
- Per-prompt observations are captured in the evaluation plan.
- Decision evidence brief exported in Markdown.
- No 'recommendation' or 'winner' section in the brief.
Caution: No persistence — the checklist resets on every visit. Capture progress in your own notes.
Evidence routes
What this kit does not promise
- Pick the right model for any product.
- Provide a live quote — pricing rows are sourced references with a retrievedAt date.
- Predict ROI or feature adoption.
- Rank vendors by price or speed.
What workflow kits do not promise
- No model recommendations or rankings. Kits walk the evidence; the reader's team makes the decision.
- No live pricing quotes. Pricing rows referenced inside the kit are sourced references with retrievedAt dates.
- No production-readiness guarantee. The kit ends with an external test plan — running those tests is the team's responsibility.
- No compliance certification, legal advice, or vendor endorsement.
- No SEO ranking guarantees or automation reliability guarantees.
- No accounts, no progress tracking, no course-completion certificates. Completion is the Markdown artifacts the kit puts in your hands.