Onboarding
Start here
Choose a role, a goal, or an artifact to find the right learning path, lab workflow, resource set, or kit. Start Here is a navigation aid — no accounts, no quiz scoring, no client-side state, no model recommendations.
Three ways to start
By role
Choose your role
Each role opens a Start Here page with a three-minute orientation, a first route list, and the artifacts the role produces. Picking a role is a navigation choice, not a quiz answer — no personalization runs against you.
Role · Newcomers and curious readers
Beginner
New to AI model selection. Start with plain-language lessons, build a first shortlist, then look at sources before considering any integration work.
First step: Walk the beginner learning path. · ~90 min
Open the role path →
Role · Engineers preparing an integration
Developer
Preparing an integration. Walk the developer path, then open the developer model evaluation kit to bundle lessons, exercises, lab playbook, prompt sets, and templates into a single Markdown work document.
First step: Walk the developer learning path. · ~180 min
Open the role path →
Role · Product managers + technical buyers
Product
Framing a model decision with engineering, finance, and legal. Walk the product manager path, then open the product model selection kit to package the review document.
First step: Walk the product manager learning path. · ~150 min
Open the role path →
Role · Automation builders, SEO operators, technical consultants
Automation
Wiring an AI model into an unattended automation. Walk the automation-specialist path, then open the automation workflow testing kit to package shadow runs, canary suite, and runbook.
First step: Walk the automation specialist learning path. · ~200 min
Open the role path →
Role · Risk, compliance, governance reviewers
Governance
Preparing an internal review for a model already in production. Walk the governance path, then open the governance review kit to package source-freshness audit, lifecycle review, and regression suite.
First step: Walk the governance learning path. · ~200 min
Open the role path →
By goal
Or — if you already know your task
Each card opens a pre-filtered view of the resource finder. No accounts, no client state — filtered URLs are noindex,follow.
Learn the basics
Plain-language concept lessons about how AI model fields behave — context, output limits, pricing references, lifecycle, hosting, status.
Open filtered view →
Choose model candidates
Build a source-backed shortlist using verified catalogue fields. Output is a /select URL anyone can re-open.
Open filtered view →
Test model behaviour
Run prompt + structured-output + regression tests in your own harness using the lab playbooks and prompt sets.
Open filtered view →
Build an automation workflow
Test an AI model inside an unattended automation before it runs in production. Shadow runs, canary suite, runbook.
Open filtered view →
Prepare a governance review
Audit source freshness, lifecycle, refusal-boundary behaviour, and package a Markdown governance review brief.
Open filtered view →
Create evidence
Package the decision brief or evaluation plan. The platform produces evidence, not recommendations.
Open filtered view →
By artifact
Or — if you already know what to produce
Each card opens a filtered finder view that lists every resource that helps produce the artifact.
Shortlist
A /select URL that captures a source-backed shortlist your team can re-open.
Open filtered view →
Comparison
A /compare/build URL that renders up to four candidate models against verified fields.
Open filtered view →
Decision brief
A paste-ready Markdown evidence pack listing verified fields, data gaps, source trail, and hosted availability.
Open filtered view →
External test plan
A short written test plan that pairs the catalogue evidence with the workload-specific tests the team will run.
Open filtered view →
Prompt test matrix
A Markdown scaffold for capturing per-candidate prompt observations without collapsing them to a score.
Open filtered view →
Source freshness checklist
A JSON or Markdown checklist of citations due for re-check, scoped per provider.
Open filtered view →
Workflow kit
A single Markdown work document that bundles lessons + exercises + lab playbook + prompt set + template for the role.
Open filtered view →
How Start Here works
- No accounts, no sign-in, no quiz scoring. Picking a role opens links — it does not save state about the reader.
- No progress tracking. There is no completion bar, no certificate, no badge. Completion is the artifacts the path puts in your hands.
- No personalization. Two readers who click the same card see the same content; no recommendation engine runs against the reader.
- No model recommendations. The role pages route into learning workflows; which model fits the workload is the reader's decision after the testing step.
- No SEO ranking guarantees, no production-readiness guarantees, no compliance certification. Start Here is a navigation aid, not a promise about outcomes.