Flui Technologies is a Constellation Software Inc. (CSI) portfolio company — Operating Group: Jonas (Vesta).
**Flui partners with raia to design and deploy a next-generation AI Support Agent, powered by a unified Core Knowledge Model and integrated with ServiceNow, to scale support for its utility-sector software infrastructure and services.**
### Snapshot (planned targets)
| Metric | Before raia | After raia (target) |
| --------------------------- | ----------------------------------------------- | -------------------------------------------------------------------------------------------------------------------- |
| First-response time | Baseline to be validated during discovery/pilot | Near-instant responses for common Tier-0 queries, 24/7 (target set after baseline review) |
| Case resolution | Baseline to be validated during discovery/pilot | Tier-0/1 deflection + faster Tier-2 resolution via better triage (targets set after baseline review) |
| Manual case-handling effort | High manual intake/triage/routing | Reduced manual effort through AI-led intake, pre-diagnosis, and guided resolution (target set after baseline review) |
| Support team focus | Repetitive Tier 0–2 tasks consume expert time | Humans focus on complex escalations and exception handling |
| Workflow consistency | Varies by agent/tier | Standardized workflows across technical, functional, and operational tiers |
**Engagement summary:** raia is partnering with Flui to deploy a Tier 0–2 AI Support Agent inside ServiceNow, grounded in a unified Core Knowledge Model built from Flui’s documentation, workflows, and historical tickets.
### 1. Company & Challenge
Flui Technologies is a leading solution provider to the **utility sector** since 1990. They specialise in delivering innovative and scalable software infrastructure and services that facilitate **market competition** and **smart metering**—critical capabilities for the transition toward a **net-zero future**.
Supporting utility-sector customers and smart-metering workflows requires fast, consistent responses across technical and operational scenarios. As support demand grows, Flui faced three compounding challenges:
1. **Manual case-handling at scale** — agents spend significant time on intake, triage, routing, and repeated “how-to” questions.
2. **Slower resolutions** — agents often need to search across documentation, workflows, and historical ticket context to reach an answer.
3. **Inconsistent workflows** — support quality and process can vary across tiers, creating operational overhead and uneven customer experience.
With complex deployments and high-stakes operational environments, Flui needed a support experience that is **fast, consistent, and knowledge-grounded**—without increasing headcount.
Flui needed an always-available system that could **understand Flui’s product ecosystem, documentation, workflows, and historical ServiceNow tickets**—providing Tier 0–2 assistance while cleanly escalating edge cases to humans.
### 2. Why Flui chose raia
Flui selected raia to move beyond basic automation and build a scalable support foundation:
* **Unified Core Knowledge Model** that consolidates knowledge from documentation, workflows, and historical ServiceNow tickets into a reusable architecture.
* **Deep ServiceNow integration** so support teams can operate within their existing system of record instead of managing parallel tooling.
* **Tier 0–2 design** that automates the repeatable majority while enabling high-quality handoffs to human experts for complex escalations.
* **Long-term AI foundation** designed to support broader AI adoption across departments over time.
### 3. Solution Design
The solution centers on an AI Support Agent built on a unified Core Knowledge Model and integrated directly into ServiceNow.
| Component | Role | Key Integrations |
| ------------------------------ | ------------------------------------------------------------------------------------------------------------------------------------------ | --------------------------------------------------------------------------------- |
| **AI Support Agent** | Handles Tier 0–2 assistance: answers common questions, guides troubleshooting, captures structured context, and proposes next steps. | ServiceNow (cases/workflows), Core Knowledge Model |
| **Core Knowledge Model** | Ingests and structures Flui’s support knowledge so the agent can retrieve accurate, policy-aligned answers and apply consistent workflows. | Product documentation, internal workflows/runbooks, historical ServiceNow tickets |
| **Agent handoff & escalation** | Escalates complex cases with full conversation history, captured context, and an AI-generated summary so humans can resolve faster. | ServiceNow Agent Workspace / routing rules |
#### Example support flows (illustrative)
* **Smart metering configuration guidance:** answer common setup questions, provide step-by-step guidance, and link to the correct internal documentation.
* **Market competition / integration troubleshooting:** guide diagnosis using known workflows and past resolution patterns from historical tickets.
* **Known issue lookup:** identify likely causes and recommended fixes based on similar ServiceNow cases.
* **Escalation with full context:** hand off complex or low-confidence cases with a structured summary, extracted signals, and recommended next steps for the support agent.
#### Deployment timeline (illustrative)
> Final phases and timing will be defined in the SOW and rollout plan.
1. **Phase 1 — Knowledge ingestion & normalization**\
Aggregate docs/workflows + historical tickets; normalize content for high-precision retrieval.
2. **Phase 2 — Pilot / shadow mode**\
AI drafts responses and triage recommendations; humans approve and provide feedback to reduce failure modes.
3. **Phase 3 — Controlled rollout**\
Expand coverage by category with monitoring, escalation policies, and continuous improvement loops.
4. **Phase 4 — Scale & standardize**\
Broaden Tier-2 assist coverage and operationalize reporting for measured outcomes.
### 4. Expected Impact (targets; measured results to follow)
Flui’s target outcomes for the engagement include:
* **Reduced manual case-handling effort** through automated intake, triage, and guided resolution (targets set after baseline review).
* **Faster time-to-resolution** by grounding responses in Flui-specific knowledge and standardized workflows (targets set after baseline review).
* **Higher consistency across tiers** via shared playbooks embedded into ServiceNow workflows (targets set after baseline review).
* **Improved team leverage** by shifting human effort toward complex escalations and exceptions.
### 5. Lessons for Prospective Clients
1. **Start with knowledge architecture, not just a chatbot.** A unified model prevents “AI sprawl” and makes improvements reusable.
2. **Meet teams where they work.** ServiceNow-native workflows drive adoption and reduce operational friction.
3. **Design for escalation, not perfection.** The highest ROI comes from handling the repeatable majority while enabling great human handoffs.
### 6. What’s Next
Flui and raia plan to expand beyond reactive support into:
* **Proactive support** (surfacing best practices, predicting issues, and reducing avoidable tickets).
* **Workflow expansion** across additional support categories and internal teams.
* **Broader AI adoption** using the same Core Knowledge Model foundation across departments.
### 7. Key Takeaway
Flui is building a **ServiceNow-integrated AI Support Agent** powered by a **unified Core Knowledge Model**—designed to deliver Tier 0–2 assistance, reduce manual case-handling, accelerate resolution time, and standardize workflows across support tiers, while creating a reusable foundation for long-term AI adoption.
---
# Agent Instructions: 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:
```
GET https://docs.raiaai.com/use-cases/case-study-flui-technologies-csi.md?ask=
```
The question should be specific, self-contained, and written in natural language.
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.