Problem
Circular-economy and compliance workflows need customer context, lead handling, follow-up, routing and owner visibility across inbound and outbound conversations.
Circular economy / customer operations
Circulogy AI employee workflows
Circulogy work connects Alex and Priya to real AI employee workflows: outbound growth and inbound customer operations supported by memory, retrieval, routing, follow-up, WhatsApp-style interaction patterns, qualification and governed escalation where supported by implementation evidence.
01
Channels
WhatsAppEmailWebAPIsVoice-ready architecture
02
Identity and context
TenantUserWorkflowBusiness state
03
AgentBus / Coordinator
RoutingAgent selectionHand-offsEscalation
04
AI employees
PriyaAlexAnika
05
MemoryOS + Agentic RAG
FactsPoliciesCorrectionsKnowledge retrieval
06
Governed execution
RBACTool permissionsHITLDecision receipts
07
Reliability
RetriesFallbacksTimeoutsObservabilityEvaluation
What was built
- Alex-style outbound/growth workflow architecture
- Priya-style inbound/customer workflow architecture
- Memory-backed context and knowledge retrieval patterns
- Inter-agent hand-off and escalation concepts
- WhatsApp/customer interaction workflow patterns
- Qualification, routing and follow-up workflows without invented performance metrics
Technologies
- Python
- WhatsApp workflows
- MemoryOS
- Supabase references
- APIs
Capabilities demonstrated
- AI Employees
- Agent hand-offs
- WhatsApp AI
- Memory
- Customer operations
- Sales automation
Sales / outbound AI employee
Alex
Alex represents outbound and growth workflows: prospect context, retrieval, follow-up, tool/API execution, governed hand-offs, and business memory where the implementation supports it.
- Prospect context
- Outbound workflow
- Persistent business context
- Tool/API execution
- Agent hand-off to inbound workflows
Evidence: Circulogy sales and growth workflow work
Customer / inbound AI employee
Priya
Priya represents inbound customer workflows: customer context, routing, WhatsApp-style interaction, acknowledgements, owner alerts, escalation, and memory-backed response logic where supported by implementation history.
- Inbound conversations
- Customer context
- Routing and escalation
- WhatsApp workflow patterns
- Hand-off to sales or owner review
Evidence: Circulogy inbound and customer workflow work
Operations / management AI employee
Anika
Anika is the clearest production-grade MarketIQX agentic workflow reference: email intake, classification, draft generation, Gmail-style tool integration, approval gates, edit/reject feedback, business memory, and learning from corrections where verified.
- Email intake
- Classification
- Draft generation
- Send / edit / reject approval
- Business memory
- Outcome correction loop
Evidence: Operational email workflow and approval-gate work
Memory and governance
Persistent context with approval and audit boundaries.
MarketIQX does not present unconstrained autonomy. The model is memory-backed, tool-aware, and governed by permissions, risk checks, and human approval where needed.
EventIdentity + ContextRecallFacts / Policies / HistoryAgent reasoningActionOutcome / CorrectionWrite-back
AI prepares action
Risk and confidence check
Low riskExecute within permission
Higher riskHuman approval: send, edit, or reject
Record outcome and correction
Related pages