# Amazon — Building an Agentic AI Platform on Amazon Bedrock AgentCore

- Company: Amazon (amazon.com)
- Announced: 2026-10-28T18:28:26+00:00
- Category: not stated
- Coverage: not counted
- Announcement: no
- Group: routine
- Source: https://aws.amazon.com/blogs/machine-learning/building-an-agentic-ai-platform-on-amazon-bedrock-agentcore/
- Record: https://forck.live/items/28903-building-an-agentic-ai-platform-on-amazon-bedrock-agentcore
- Subject: Bedrock / Nova

Amazon's post walks through the architecture and a reference implementation for an enterprise agentic AI platform on Amazon Bedrock and Bedrock AgentCore, aimed at platform teams supporting many agent-building teams. It argues the model and tool use are not the blocker; the bottleneck is everything around them, such as which team owns each model and tool and how cost and health stay visible across accounts. The design uses four connected planes — developer experience, platform control, Workstream execution, and assurance and operations — and scales by a Workstream cell rather than a manually configured agent or one shared workload account. Separate inference and tool gateways govern access: every untrusted inference request passes an approved gateway that applies guardrails, restricts models and Regions, and enforces per-team rate limits, token quotas and budgets. The blueprint, built on AgentCore, adds ownership, account boundaries, delivery, observability, cost governance and retirement as deployable infrastructure as code, and the post says readers can deploy a governed Workstream from it. The blueprint implements a federated operating model, which the post says is where most enterprises land, with a central platform team providing governed services while domain teams own their agents, tools and data.

## Evidence

Verbatim from https://aws.amazon.com/blogs/machine-learning/building-an-agentic-ai-platform-on-amazon-bedrock-agentcore/:

> If you’ve tried to get an agent into production, the model and tool use are not the blocker. The main bottleneck is everything around them: how many teams can build within a secure environment, which team owns each model and tool, how AgentOps and error handling stay inside a pre-defined boundary, and how platform, Workstream, and finance teams see cost and health across accounts. For a Platform team that wants to support many agentic use cases, that’s a platform product to design and operate. In this post, we refer to the team that manages the platform as the Platform team, and to the teams that build on it as Workstream teams. Workstream teams are the domain or product teams that build and own agents on the platform.
> 
> Each Workstream team needs a paved road with approved models and tools, short-lived workload identity, isolated runtime and memory, automated evaluation, rollback, telemetry, and cost attribution. These controls need to remain consistent as Workstream teams customize prompts, tools, data access, and application behavior. When every team assembles them, controls drift, shared services become a bottleneck, and ownership is difficult to audit.

---

Record: https://forck.live/items/28903-building-an-agentic-ai-platform-on-amazon-bedrock-agentcore
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md