Assess → Build → Operate
Three phases. One accountable team. Monthly outcome reports that prove the work.
Phase 1
The Assessment
The Assessment is a paid diagnostic — not a sales call. Your embedded engineer runs a structured analysis of your environment over 1–2 weeks and delivers a four-section report covering automation exposure, what's ready to activate, custom candidates, and recommended path forward.
Current Automation Exposure
Which systems are connected, estimated monthly automatable events, and your current coverage percentage — what's handled today vs. what's falling through.
Ready to Activate
Automation that works against your stack today: what's already running, what can go live immediately, and the expected impact of each.
Custom Automation Candidates
Workflows specific to your business that fall outside the standard catalog — prioritized by effort, value, and how closely they match patterns we've already built.
Recommended Path Forward
Tier recommendation (Starter, Standard, or Growth), which outcomes to activate first, and the sequencing that gets you to verifiable results fastest.
Phase 2
The Custom Build
Based on the Assessment report, your embedded engineer models your operations as a semantic model and builds custom AI workflows on top using generative AI tools (Claude, Codex). This is not template selection — it's implementation against your thresholds, your escalation paths, your edge cases.
- ✓ Custom AI built on your semantic model — executes deterministically against your business rules, no hallucinations, full audit trail
- ✓ Connects to the system categories your operations already run on — no rip and replace
- ✓ Reliability built-in: deduplication, retry, dead-letter queue, full execution audit trail
- ✓ Typically 2–6 weeks from Assessment to first live outcome, depending on integration complexity
Phase 3
The Embedded Engineer
Your dedicated embedded engineer runs the platform on your behalf — indefinitely. This is the part most teams underestimate. Running AI in production isn't shipping a model: it's modeling your operations, threshold tuning, handling edge cases as your business evolves, and monthly proof that it worked.
Embedded engineer continuity: Your semantic model, execution history, and business logic live in the platform — not with any individual. If your embedded engineer ever changes, continuity is built in. No re-onboarding.
Run the Assessment
Run a structured analysis of the client's environment over 1–2 weeks and deliver a four-section report covering automation exposure, what's ready to activate, custom candidates, and recommended path forward.
Build custom AI on your semantic model
Build custom AI automation against your business rules using generative AI tools (Claude, Codex). Your operations become a semantic model the AI executes deterministically against — no hallucinations, full audit trail.
Monitor and tune
Review execution logs, adjust conditions as the client's operations evolve, and activate new skills from the catalog as they ship.
Deliver monthly outcome reports
Measure verified results against the agreed baseline and deliver a clear breakdown of outcomes produced each month.
Manage the engagement
Handle credential setup, integration changes, and tier transitions — so clients never need to maintain an internal automation team.
Concierge pricing
Concierge tiers, outcome-aligned
Base retainer covers your embedded engineer's time. The outcome fee means we earn more only when your business produces more — aligned by design.
Starter
$750/mo
base retainer / month
- Outcomes / month 1 outcome / month
Validating your first outcome before scaling embedded engineering across operations.
Most common
Standard
$1,500/mo
base retainer / month
- Outcomes / month Up to 3 outcomes / month
Operations teams ready to run three production outcomes against their support and fulfillment workflows.
Growth
$5,000/mo
base retainer / month
- Outcomes / month Up to 10 outcomes / month
Operations leaders running embedded engineering across multiple workflows — at a fraction of a single in-house ops engineer.
All tiers include an outcome fee scoped to your operations and agreed during the Assessment — so the incentive is always aligned with your results.
Not sure which tier fits? Your Assessment report includes a tier recommendation based on your stack and event volume.
Book your Assessment →Why we charge for outcomes, not operations
Most automation platforms charge per task, per seat, or per API call — regardless of whether those operations produce results. That creates a fundamental misalignment: the platform earns more even when your business gets nothing.
Clarissi's hybrid model (base retainer + outcome fee scoped to your operations and agreed during the Assessment) aligns us to your business outcomes. Your operator earns more only when you produce more. Enterprise analysts predict this is where AI pricing is heading — we're already there.
What we don't do
Clarissi is embedded engineering for AI-native operations. It's not the right fit for everyone — and we'd rather be clear about that upfront.
- ✗ No self-serve signup. Every engagement starts with a paid Assessment so we can model your operations correctly and build the right outcomes, not a generic template.
- ✗ No per-seat SaaS licensing. We charge for outcomes, not access.
- ✗ No DIY workflow builder. If you want a drag-and-drop canvas to build your own automations, there are better tools for that.
- ✗ No engagements without an Assessment. We don't activate outcomes without modeling your operations first. The Assessment is what makes the build defensible.