Scopes of work

How an idea becomes a running system.

Five scopes, run in sequence or à la carte. Each one ends in something real: a map, a working prototype, a shipped system, a trained team, or a system kept healthy.

Scope 01

AI-Native Discovery

We find the single workflow where AI changes the economics, then map exactly what it takes to ship it.

Deliverables

  • Workflow and data audit
  • AI-native opportunity map
  • Scoped recommendation with effort and impact
Scope 02

Proof of Concept

Prove the highest-value workflow on your real data, and get an honest go or no-go before committing to a build.

Deliverables

  • Working prototype on live data
  • Evaluation against real cases
  • Go / no-go recommendation
Scope 03

Phased Delivery

Production AI shipped in milestones, with Claude in the loop for the work that actually recurs.

Deliverables

  • Production system on your stack
  • Agentic and scheduled workflows
  • Observability, guardrails, and spend caps
Scope 04

Enablement & Workshops

Hands-on sessions that get your team building with Claude Code and agents, not just watching demos.

Deliverables

  • Tailored workshop for your domain
  • Reusable skills and agent templates
  • Playbooks your team keeps
Scope 05

Managed Retainer

We keep the system healthy, on call, and improving, so the agents you shipped stay reliable.

Deliverables

  • Monitoring and incident response
  • Iteration on prompts, models, and flows
  • Optional coaching add-on
Let's talk

Not sure which scope you need?

Most engagements start with one painful, recurring workflow. Tell me yours and I'll point to the right starting scope.

brandon@aurapathai.com
Start a conversation

Have a workflow worth automating?

The best engagements start with one painful, recurring workflow and a willingness to put real data behind it. If that sounds like you, let's talk.