> For the complete documentation index, see [llms.txt](https://docs.duplocloud.com/docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.duplocloud.com/docs/extension-framework/case-studies/salesops-platform.md).

# AI-Powered SalesOps

To demonstrate that the Extension Framework is domain-agnostic, a second application was built on the same CaaS platform: an AI-powered demand generation platform for sales teams. It uses identical framework mechanics — resource taxonomy, ticket-based execution, cost tracking per resource, RBAC inherited from workspaces — with a completely different domain. There is no infrastructure, no cloud provider, no Kubernetes. There are prospects, campaigns, qualification signals, and outreach sequences.

## The Policy Model

* **Demand Gen** — the top-level container for all sales operations activity.
* **Cohort** — a batch of target accounts grouped for a campaign run (e.g., "May\_02"). Tracks total cost across all qualification and engagement activity within it.
* **Account** — a prospective company. The spec is intentionally minimal: company name and website. The framework and skill handle the research from there.
* **Qualification** — AI-powered research and fit analysis. Created automatically when an Account is added to a Cohort. The agent evaluates the account against ICP criteria and returns a structured verdict.
* **Engagement** — outreach execution. The agent selects a sequence template from a connected sequencing provider (Outreach or Lemlist) and enrolls qualified leads. Engagement depends on a completed Qualification — the same dependency enforcement mechanism used in DevOps.

## How Qualification Works

When a user adds an account to a cohort, a Qualification resource is created automatically — no separate action required. The associated skill contains the organization's ICP criteria. The agent researches the prospect using web search and the Apollo provider, evaluating multiple signals in parallel:

* **Cloud Infrastructure** — technologies detected on the company's infrastructure (e.g., AWS, Amazon SES, Route 53, CloudFlare Hosting) → Verdict: PRESENT
* **DevOps Help Needed** — analysis of open job postings; example: 0 DevOps roles hiring → Verdict: LOW
* **Funding** — funding history and recency; example: Series B $31,500,000 (2024), Tier: STRONG → Verdict: STRONG

The result includes an Overall Verdict, a list of extracted leads with contact details, and the full signal breakdown. The ticket sidebar shows the running cost ($0.636 in the example above), company metadata (Industry, Location, Employees), and current status — the same sidebar layout used for every resource across every extension.

## How Engagement Works

For accounts that pass qualification, the user initiates the Engagement stage. A multi-step wizard collects configuration across four steps:

1. **Sequence** — select a source sequence from a connected sequencing provider (Outreach or Lemlist).
2. **Email & Sender** — choose the sending identity and reply-to configuration.
3. **Leads** — confirm which qualified leads to enroll, drawn from the Qualification result.
4. **Trigger** — configure when and how the sequence activates.

The agent then executes the sequence through the connected provider. Like every other resource in the framework, this execution is backed by a ticket — observable in real time, linked from the Engagement detail view, and included in cost tracking.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.duplocloud.com/docs/extension-framework/case-studies/salesops-platform.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.