AI & Cloud Infrastructure

Commerzbank's Enterprise AI Agents at Scale with Microsoft Foundry - Microsoft Ignite 2025

By Technspire Team

•

November 28, 2025

•

5267 views

Use Case Identification (2-3 weeks)

• Identify high-value agent opportunities (customer service, operations automation)
• Define agent capabilities and scope (what can it do? what's out of scope?)
• Map required integrations (what systems must agent access?)
• Establish success metrics (response time, resolution rate, user satisfaction)
• Prioritize 2-3 pilot agents for initial deployment

Foundry Environment Setup (3-4 weeks)

• Deploy Microsoft Foundry in Azure subscription
• Configure Entra Agent ID integration
• Set up AI Gateway with private network connectivity
• Configure Foundry Control Plane for monitoring
• Establish development, staging, production environments
• Integrate with Microsoft 365 tenant

Custom Tool Development (4-6 weeks)

• Build custom tools for business-specific systems
• Implement authentication and authorization
• Register tools in Foundry tool catalog
• Test tools with sample data and scenarios
• Document tool capabilities for agent developers
• Establish tool versioning and update process

Pilot Agent Development (6-8 weeks)

• Build 2-3 pilot agents using Microsoft Agent Framework
• Implement agent logic (conversation handling, tool invocation)
• Configure Entra Agent IDs with appropriate permissions
• Test agents thoroughly (unit tests, integration tests, user acceptance)
• Deploy to staging environment with limited user group
• Gather feedback and iterate on agent capabilities

Production Deployment (4-6 weeks)

• Gradual production rollout (canary deployment)
• Monitor closely (errors, performance, user satisfaction)
• Integrate agents into Microsoft 365 (Teams, Outlook, SharePoint)
• Train employees on using agents effectively
• Establish support process (how users report issues)
• Document agent capabilities and limitations

Scale and Optimize (Ongoing)

• Expand agent fleet based on pilot success
• Optimize costs (model selection, prompt engineering, caching)
• Enhance agent capabilities based on user feedback
• Monitor Foundry Control Plane dashboards weekly
• Measure ROI (time saved, user satisfaction, cost reduction)
• Evolve governance as agent ecosystem matures

The Future: Multi-Agent Networks and Cross-Company Collaboration

The BRK186 session concluded with a vision of the next evolution: agents collaborating across organizational boundaries.

Multi-Agent Orchestration

Complex business processes require multiple specialized agents working together:

Example: Enterprise Procurement Workflow

PurchaseRequestAgent: Employee requests new equipment, agent validates budget availability
ApprovalAgent: Routes request to appropriate managers based on amount, tracks approvals
VendorAgent: Searches vendor catalog, compares prices, negotiates quotes
ComplianceAgent: Verifies vendor meets organizational standards (security, sustainability)
OrderAgent: Places order, tracks shipment, updates inventory system
InvoiceAgent: Receives invoice, matches to purchase order, processes payment
AssetAgent: Registers asset in asset management system, assigns to employee

Coordination: All agents orchestrated through Foundry—each knows its role, hands off to next agent, logs every decision for audit.

Cross-Company Agent Collaboration

Future vision: Your agents can securely interact with partner/customer agents:

Supply chain: Your inventory agent talks to supplier's fulfillment agent—automated ordering, shipment tracking
Customer service: Your support agent collaborates with customer's IT agent to diagnose issues
Financial services: Your payment agent interacts with bank's clearing agent for instant transactions
Healthcare: Hospital's patient agent securely communicates with pharmacy's prescription agent

Security foundation: Entra Agent ID extends to external agents. Your agent verifies partner agent's identity before sharing data. All cross-company interactions logged for audit. Data governance enforced (what can be shared, with whom, under what conditions).

Key Takeaways from BRK186

✓ Commerzbank's Ava handles thousands of customer requests monthly using Microsoft Foundry
✓ Microsoft Entra Agent ID provides unique identities, authentication, RBAC, and audit trails for agents
✓ Foundry AI Gateway enforces security guardrails, rate limiting, and prevents prompt injection attacks
✓ Full lifecycle management: Development → testing → staging → production → updates with governance
✓ Custom tool integration: Connect agents to your business systems via REST APIs in tool catalogs
✓ Microsoft 365 integration: Agents appear in Teams, Outlook, SharePoint as digital coworkers
✓ Foundry Control Plane provides unified governance, cost tracking, and monitoring at scale
✓ Organizations report 72-95% automation of routine tasks with enterprise agents

Deploying AI agents at enterprise scale is fundamentally different from proof-of-concept demos. Microsoft Foundry provides the identity, connectivity, security, lifecycle management, and governance capabilities that production agents require. Commerzbank's success with Ava demonstrates the power of this platform—autonomous agents handling thousands of customer interactions with the security and compliance financial services demand. For Swedish organizations ready to move beyond agent pilots, Foundry is the blueprint for secure, scalable, and governable agent deployments that deliver real business value.

Ready to Transform Your Business?

Let's discuss how we can help you implement these solutions and achieve your goals with AI, cloud, and modern development practices.

No commitment required • Expert guidance • Tailored solutions