A multi-agent ecosystem on a single Agent Builder — agents that research, debate in councils, and hand off to humans, working live across every channel.
Agent BuilderMaster OrchestratorCouncil of AIHuman-in-the-loopSelf-learning
01
Challenge 01
Housing-Loan Arrears Rescheduling
Intake · parallel
Read the case
Case Intake✓
Policy✓
Risk & Hardship✓
Decision
Compliant plan
20% deduction ceiling
within original term ✓
Council · debate
Three advisors + chair
Approve
Approve
w/ cond.
Human-in-the-loop
Tiered sign-off
Auto-approve
Officer review
Refer to human
Builder agents run the whole case — parallel intake → compliant plan → council debate → tiered human sign-off. Cached, with openable attachments.
02
Challenge 02
Country Intelligence Briefings
Research · parallel
Live, multi-source
Profile✓
Economy & Energy✓
UAE Opportunities✓
Risk & Outlook✓
Strategic Council
Advisors debate + skeptic
Pursue
Pursue
w/ cond.
Devil's adv.
Executive report
Verdict + AI slide deck
82%confidence
talking pointsslides
Four agents gather live data → a Strategic Council debates → one decision-ready briefing, grounded to today, with an AI-generated slide deck.
03
Challenge 03
Omnichannel AI Engagement
Website
WhatsApp
Phone
Master Orchestrator
Routes to any specialist agent
Citizen Assistant
Housing Eligibility
Arrears Rescheduling
Licensing
Self-Service Knowledge
Country Intelligence
Officer Co-Pilot
Leadership Dashboard
Hand off to a human officer
Tawasul · Citizen
How do I apply for housing support?
You'll need your Family Book & Emirates ID — I can start it now.
Can you check my arrears too?
● handled by Housing agent
Citizen portal · web · WhatsApp · phone
Officer Co-Pilot
Ahmed Al Mansoori high risk
Next best action Confirm income docs, offer a reschedule plan.
“Thank you, Ahmed — based on your salary certificate you qualify for…”
⚠ escalate to a human officer
Co-Pilot for the service officer
Command Center
1,240cases
82%CSAT
54%auto-resolved
Auto-resolved vs handed off
Leadership dashboard · AI-generated
One master orchestrator powers all three surfaces — routing, the officer Co-Pilot, and AI-generated leadership dashboards — learning from every resolved case.
The Agent Builder
Compose anything — no code
KnowledgeGraph RAG
Ingest any source
PDFWordExcelCSVTXTMDImageVideo URLWeb linkOpen Data
Attached to the agent
✓Housing policyMD
✓Tariff scheduleXLSX
✓Bayanat datasetOPEN DATA
Docs, media, web & datasets — chunked, embedded into pgvector & linked as a knowledge graph (GraphRAG)
Agent31 built
Persona
“You are the Housing Eligibility Agent — assess eligibility, cite the rule…”
Model
Google Vertex AI
Tools granted
policy_searchapplicant_lookupweb_search
Bound service
Housing Assistance
Persona + model + knowledge + tools, bound to a service
Connector7 wired
Kind
HTTP APIMCPOpen DataBayanat
Endpoint
api.worldbank.org/v2/country/{ISO3}/indicator…
World BankHTTP
DeepWikiMCP
MOEI Open DataCKAN
REST APIs, MCP servers & open-data portals — granted to agents as tools
Workflow6 built
Ordered stages
Intakeparallel ×3
Decisionsingle
Councildebate ×4
Synthesissingle
Chain stages that run agents in parallel, alone, or as a debating council
Author knowledge, agents, connectors and workflows in the Builder — parallel, single or council stages. Every surface and the orchestrator use them live.