MRC Talking Points - Modern Agentic Applications Day
Core thesis: "Enterprises that succeeded at digital transformation also do Agentic AI well."
1. 30-Second Elevator Pitch (for phone)โ
AI agents are changing how enterprises work. But AI agents do not fall from the sky โ they require robust cloud-native infrastructure as a foundation. Serving LLMs with GPU Operator and vLLM on EKS and operating agents with Bedrock AgentCore and Hosted MCP is the practical answer. At Modern Agentic Applications Day on April 9, see everything from cost optimization to AgentOps in a single day. Limited to 200 attendees.
2. Two-Minute Summary (for meeting openers)โ
In 2026, 80% of enterprises plan to adopt AI agents. But most companies overlook one fact โ AI agents are a combination of dozens of microservices (LLM serving, vector DB, gateway, monitoring). Production operation is impossible without container orchestration.
AWS addresses this challenge on two tracks:
-
Modern Agentic Infra: EKS is the managed Kubernetes that fully supports LLMOps. With GPU Operator and run.ai you manage GPU resources efficiently, and with vLLM + llm-d you serve your own LLMs. Combining MIG partitioning with Karpenter autoscaling achieves cost optimization and operational stability at the same time. With Hybrid Nodes, on-premises is also unified under EKS, operable at about $4,900/month at 50 req/s.
-
Modern Agentic Apps: A rich managed-agent ecosystem from AWS is available. Bedrock AgentCore operates the agent runtime as a managed service, Hosted MCP servers standardize agent-to-tool integration, and managed agents such as DevOps Agent automate infrastructure operations. Migrating 15+ self-hosted Pods to managed services reduces operational complexity by 80%.
At Modern Agentic Applications Day on April 9, see both tracks' deep-dive sessions and real-world build cases in a single day. After the event, Modern Agentic Infra / Modern Agentic Apps PoCs across two tracks can start immediately. Capacity is capped at 200, so early registration is recommended.
3. Industry-Specific Hooksโ
E-commerceโ
- AI recommendation agents deliver real-time per-customer personalization โ 35% conversion uplift case
- GPU MIG + Semantic Caching operates at the level of $4,900/month at 50 req/s
- For companies that have completed digital transformation (containers), AI transformation is the natural next step
Financial Services (FSI)โ
- Data sovereignty and regulatory compliance are key โ Hybrid Nodes protect on-premises data while leveraging the EKS AI platform
- Build real-time fraud detection agents, achieving 99.2% detection accuracy
- Safe AI operations even in financial regulatory environments โ Pod Identity, encryption, and audit logs fully in place
Gamingโ
- AI NPC agents deliver real-time conversational experiences โ 2.5ร increase in playtime
- vLLM + llm-d delivers low-latency inference serving (TTFT shortened by 40%)
- Stable operation even at millions of concurrent users via Karpenter autoscaling
Manufacturingโ
- Vision AI quality inspection agents โ 90% defect-rate reduction case
- GPU MIG partitions one GPU across 7 workloads โ 75% cost savings
- Unified EKS management even at the factory edge via Hybrid Nodes
Telcoโ
- Network optimization AI agents deliver 50% operational efficiency uplift
- Hybrid Nodes are optimal for edge AI workloads โ essential infrastructure for the 5G/6G era
- Build a unified AI platform spanning core network to edge with EKS
4. Anticipated Questions & Answersโ
Q1. What's different between AI agents and existing chatbots?โ
Chatbots handle only predefined rules or simple Q&A. AI agents autonomously use tools, make decisions, and execute multi-step tasks. For example, given "create this month's sales report," they autonomously perform data lookup โ analysis โ visualization โ report generation.
Q2. We haven't even adopted containers yet.โ
The Modern Agentic Infra track is exactly that starting point. EKS-based container migration is the first step toward AI-ready infrastructure, and Hybrid Nodes unify on-premises and cloud management. The event covers a phased migration roadmap in detail.
Q3. GPUs are expensive โ AI adoption is a financial burden.โ
That is exactly why GPU optimization is critical. On EKS, GPU Operator + run.ai precisely allocate GPUs at the workload level, and MIG partitioning divides a single GPU into up to 7 partitions, enabling 75% cost savings. Adding Semantic Caching and Karpenter autoscaling reduces costs further.
Q4. Isn't it enough to use only Bedrock?โ
Bedrock is a good starting point, but if you need self-hosted model serving (vLLM), custom agent logic, or hybrid composition, an EKS-based platform is essential. AWS's strength is the ability to combine the two โ operate the agent runtime as a managed service via AgentCore, connect first-party tools via Hosted MCP servers, and serve required models directly on vLLM atop EKS.
Q5. What are AWS's advantages over GCP/Azure?โ
GCP and Azure offer similar services. AWS's differentiator is that the full LLMOps stack is integrated on EKS. GPU Operator, run.ai, vLLM, and Karpenter compose the model-serving infrastructure, and AgentCore + Hosted MCP + DevOps Agent automate agent operations. EKS is also the only one to offer a 99.99% SLA dedicated control plane (PCP).
Q6. What can I get from the event?โ
You can see practical guides on AI infrastructure architecture and agent applications all in one day. Capacity is 200, so CTO/VP-level networking opportunities are also provided. Two tracks are ready to start a PoC immediately after the event.
Q7. How does the PoC proceed?โ
Two tracks are prepared:
- Modern Agentic Infra: EKS + GPU Operator + vLLM PoC โ 2-week PoC โ production roadmap
- Modern Agentic Apps: Bedrock AgentCore + Hosted MCP PoC โ agent build and ops automation
Event attendees receive PoC priority.
Q8. How are security and regulatory concerns addressed?โ
EKS Pod Identity for fine-grained access control, Hybrid Nodes for data sovereignty, and Private Subnets + VPC isolation for network security. Build cases under financial/healthcare regulatory environments are shared at the event.
Q9. We already have a Kubernetes cluster โ can we add AI workloads?โ
Yes. Migrating to EKS Auto Mode enables automatic GPU node provisioning, and existing workloads and AI workloads can coexist on the same cluster. Karpenter automatically allocates GPU instances suited to AI workloads.
Q10. Who is the event audience?โ
The primary audience is CTOs, VPs of Engineering, platform engineering leaders, and IT decision makers. It is most suitable for technical decision makers shaping an AI strategy or in the midst of digital transformation. The covered industries focus on E-commerce, FSI, Gaming, Manufacturing, and Telco.
5. Competitive Differentiationโ
| Comparison Item | AWS (EKS) | GCP (GKE) | Azure (AKS) |
|---|---|---|---|
| Control plane | PCP dedicated (99.99% SLA) | Shared (99.95% SLA) | Shared (99.95% SLA) |
| Data-plane automation | Auto Mode + Karpenter | Autopilot + NAP | NAP (Node Auto-provisioning) |
| AI agent runtime | AgentCore (MCP-native) | Vertex AI Agent Builder | Azure AI Agent Service |
| GPU optimization | Karpenter + MIG/DRA integrated | MIG support + GKE NAP | MIG support |
| AI-dedicated chips | Inferentia2, Trainium | TPU v5e/v5p | Maia 100 (limited) |
| Migration | Transform (K8s integrated) | Migrate to Containers | Azure Migrate |
| GPU instances | H100, L40S, Inferentia2, Trainium | H100, A100, L4, L40S | H100, A100, MI300X |
GCP and Azure offer similar services, but AWS differentiates with a fully integrated LLMOps stack on EKS (GPU Operator, run.ai, vLLM, Karpenter) and with a managed-agent ecosystem (AgentCore + Hosted MCP + DevOps Agent).
6. Follow-Up Pipelineโ
Track 1: Modern Agentic Infra (EKS + GPU + LLMOps)โ
| Stage | Content | Duration |
|---|---|---|
| Assessment | Analyze current infrastructure + define AI workload requirements | 1 week |
| PoC | Compose EKS + GPU Operator + run.ai + vLLM + Hybrid Nodes | 2 weeks |
| Production roadmap | Design production architecture + cost-optimization plan | 1 week |
Track 2: Modern Agentic Apps (AgentCore + MCP + agent development)โ
| Stage | Content | Duration |
|---|---|---|
| Agent design | Define use cases + design agent architecture | 1 week |
| PoC | AgentCore + Hosted MCP server integration + agent build | 2 weeks |
| Operations automation | DevOps Agent integration, monitoring, CI/CD setup | 1 week |
Event attendees receive PoC priority allocation and dedicated AWS SA support.
7. Event Summaryโ
| Item | Details |
|---|---|
| Event | Modern Agentic Applications Day |
| Date | April 9, 2026 (Wed) |
| Venue | Centerfield 18F |
| Capacity | Limited to 200 |
| Audience | CTOs, VPs of Engineering, platform engineering leaders, IT decision makers |
| Key sessions | Modern Agentic Infra / Modern Agentic Apps |
| Follow-up program | 2 PoC tracks (Agentic Infra / Agentic Apps) |
- "If you're curious about the next step after digital transformation"
- "If AI infrastructure cost is a concern โ we share a case of 75% GPU savings"
- "Capacity is limited to 200, so early registration is recommended"
- "PoCs can start immediately after the event"