Change
79e0893
79e0893968aa16b58827434b230d7ad449279a82 · commit on GitHub
aws-blog-feed: changed (598958 bytes, HTTP 200)
raw/aws-blog-feed/response.xml modified
- Source
- aws-blog-feed
- Lines added
- +3,912
- Lines removed
- -4,148
- Stored bytes at this commit
- 598,958
- Timestamp
- observed
- Raw artifact at this commit
- raw/aws-blog-feed/response.xml
Recorded headers
| observed_at | 2026-10-30T05:49:21.336Z |
|---|---|
| origin_date | null |
| status | 200 |
| final URL | https://aws.amazon.com/blogs/machine-learning/feed/ |
| etag | null |
| last-modified | Thu, 29 Oct 2026 23:10:12 GMT |
| date | Fri, 30 Oct 2026 05:49:21 GMT |
| age | null |
| cache-control | null |
| cf-cache-status | null |
| content-encoding | null |
| content-length | null |
@
@@ -5,7 +5,7 @@ <atom:link href="https://aws.amazon.com/blogs/machine-learning/feed/" rel="self" type="application/rss+xml"/> <link>https://aws.amazon.com/blogs/machine-learning/</link> <description>Official Machine Learning Blog of Amazon Web Services</description>-
<lastBuildDate>Wed, 28 Oct 2026 18:28:26 +0000</lastBuildDate>+
<lastBuildDate>Thu, 29 Oct 2026 20:16:40 +0000</lastBuildDate> <language>en-US</language> <sy:updatePeriod> hourly </sy:updatePeriod>@
@@ -13,183 +13,360 @@ 1 </sy:updateFrequency> <item>-
<title>Building an Agentic AI Platform on Amazon Bedrock AgentCore</title>-
<link>https://aws.amazon.com/blogs/machine-learning/building-an-agentic-ai-platform-on-amazon-bedrock-agentcore/</link>+
<title>How the UK Amazon Quick User Group grew in 2026</title>+
<link>https://aws.amazon.com/blogs/machine-learning/how-the-uk-amazon-quick-user-group-grew-in-2026/</link> -
<dc:creator><![CDATA[Omar Elkharbotly]]></dc:creator>-
<pubDate>Wed, 28 Oct 2026 18:28:26 +0000</pubDate>-
<category><![CDATA[Amazon Bedrock AgentCore]]></category>+
<dc:creator><![CDATA[Priyanka Bettaswamygowda]]></dc:creator>+
<pubDate>Thu, 29 Oct 2026 20:16:40 +0000</pubDate>+
<category><![CDATA[Amazon Quick Suite]]></category> <category><![CDATA[Intermediate (200)]]></category>-
<category><![CDATA[Technical How-to]]></category>-
<guid isPermaLink="false">8bd827344f7a730b420aced0542c5d0dd666efbf</guid>+
<category><![CDATA[Uncategorized]]></category>+
<guid isPermaLink="false">106cb441d765c495deeab0901e357357287aed51</guid>-
<description>Stand up a governed, multi-account Agentic AI Platform on Amazon Bedrock AgentCore. This post walks through the architecture and an open-source reference blueprint: the operating model, control plane, per-Workstream cells, separate inference and tool gateways, and a CI/CD evaluatio…-
<content:encoded><p>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 s…-
<p>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 ac…-
<p>In this post, we walk through the architecture and reference implementation of an enterprise Agentic AI Platform on <a href="https://aws.amazon.com/bedrock/" target="_blank" rel="noopener">Amazon Bedrock</a> and <a href="https://aws.amazon.com/bedrock/agentcore/" target="_bla…-
<p><a href="https://aws.amazon.com/bedrock/agentcore/" target="_blank" rel="noopener">Amazon Bedrock AgentCore</a> provides modular services for building, deploying, and operating agents at scale, including runtime, gateway, identity, memory, policy, registry, and evaluations. The …-
<h2 id="architectural-principles">Architectural principles</h2> -
<p>Before diving into the solution details, it helps to understand the organizational needs that drive the platform architecture. The common challenges we have heard from customers who establish an agent platform fall into these five categories:</p> -
<ol type="1"> -
<li><strong>Governed model access:</strong> Every untrusted inference request must pass through an approved inference gateway. The gateway applies the organization’s guardrails (for example, <a href="https://aws.amazon.com/bedrock/guardrails/" target="_blank" rel="noopener">…-
<li><strong>Governed tool access:</strong> Agents must discover and invoke only approved tools through an authenticated Model Context Protocol (MCP) gateway, with exact targets, least-privilege execution roles, policy decisions, and an auditable owner.</li> -
<li><strong>Delegated delivery:</strong> Workstream teams (for example, product teams) need to ship and roll back independently inside pre-approved boundaries, without waiting for the platform team to deploy each application or granting developers a direct infrastructure write pat…-
<li><strong>Fleet operations and cost:</strong> Several personas need visibility across accounts without gaining write access to team workloads. Agent operations engineers need logs, metrics, traces, and quotas. Workstream owners need cost per agent against their budgets. FinOps a…-
<li><strong>Bounded evidence and lifecycle:</strong> Every agent release needs proof at each boundary it touches: -
<ol type="a"> -
<li><strong>Model access:</strong> An approved model call succeeds, and a call to an unapproved model or Region is denied.</li> -
<li><strong>Tool access:</strong> A subscribed tool call succeeds, and a call to a tool outside the agent’s subscription is denied with the expected error.</li> -
<li><strong>Identity:</strong> A request from another workstream’s account or identity is rejected.</li> -
<li><strong>Release:</strong> The previous agent runtime version can be restored.</li> -
<li><strong>Retirement:</strong> Teardown removes the agent’s runtime, memory, tool gateway, and tool grants in dependency order, and confirms nothing is left behind.</li> -
</ol> </li> -
</ol> -
<p>The blueprint addresses these needs through four connected planes: developer experience, platform control, Workstream execution, and assurance and operations. The unit of scale is the Workstream cell, not a manually configured agent or one shared workload account. The blueprint <a href="…-
<h2 id="operating-models-and-platform-components">Operating models and platform components</h2> -
<p>Enterprises usually adopt one of three operating models for agentic AI:</p> +
<description>The UK Amazon Quick User Group's 2026 roundup: three quarterly meetups, 150+ attendees, and customer stories from Arriva, Carsa, and Propulse Lab on using agentic AI and Amazon Quick to turn data into faster decisions, higher satisfaction, and scalable growth.</description>+
<content:encoded><p>The <a href="https://www.linkedin.com/groups/16948010/" target="_blank" rel="noopener">Amazon Quick User Group – United Kingdom</a> is a community of <a href="https://aws.amazon.com/quick/" target="_blank" rel="noopener">Amazon Quick</a> us…+
<p>Whether you’re already using Amazon Quick or just getting started, we’d love for you to join the UK Amazon Quick User Group and become part of a growing community that’s redefining how organizations make decisions with data. In this post, we’re thrilled to share our 2026 yearly roundup, hig…+
<h2 id="why-amazon-quick-user-groups-matter">Why Amazon Quick user groups matter</h2> +
<p>The <a href="https://www.linkedin.com/groups/16948010/" target="_blank" rel="noopener">UK Amazon Quick User Group</a> is a local community of Amazon Quick users who come together quarterly to share knowledge, best practices, and experiences with the service. Amazon Quick user gr…<ul> -
<li><strong>Centralized:</strong> A single artificial intelligence and machine learning (AI/ML) team builds and runs every agent. Governance is consistent, but that team quickly becomes a bottleneck as the number of use cases grows.</li> -
<li><strong>Decentralized:</strong> Each business unit builds and operates its own agents. Teams can move quickly, but without shared standards, controls tend to drift, effort gets duplicated, and cost visibility suffers.</li> -
<li><strong>Federated:</strong> A central Platform team provides shared, governed services while domain teams retain ownership of their agents, tools, and data. This model balances team autonomy with consistent guardrails and is where most enterprises land.</li> -
<li>For assistance choosing, see the <a href="https://docs.aws.amazon.com/prescriptive-guidance/latest/govern-architect-agentic-ai/governance-models.html" target="_blank" rel="noopener">AWS Prescriptive Guidance page on Governance Models</a>. </li> +
<li><strong>Learn and share</strong> – Exchange real-world insights about the capabilities of Amazon Quick, from interactive dashboards and data storytelling to agentic AI-powered automations and custom agents. Members share what works, what doesn’t, and new approaches that drive …+
<li><strong>Network</strong> – Connect with regional professionals who are solving similar use cases with Amazon Quick for enterprise AI applications, data analytics, and workplace productivity.</li> +
<li><strong>Problem-solve</strong> – Tackle challenges collaboratively by discussing solutions and use cases across different industries and business contexts.</li> +
<li><strong>Stay updated</strong> – Get firsthand knowledge about new features, updates, and emerging best practices directly from peers and experts.</li> +
<li><strong>Build community</strong> – Foster a local support network that complements the broader Amazon Quick community, with trusted contacts who understand UK market and business context.</li> </ul> -
<p>Following patterns we have observed in customer engagements, the blueprint implements the federated model. The Platform team is the hub, typically a Platform product manager, architect, platform engineers, agent operations engineers, and a security and governance lead. Workstream teams are …-
<p>In this blueprint, the Platform team owns the definition of each component: templates, versions, policies, and admission. Each Workstream account runs its own instance of the per-team components (Runtime, Memory, Tool Gateway), so a failure or rollback stays inside one team’s boundary.</…-
<h2 id="solution-overview">Solution overview</h2> -
<p>This solution delivers a governed, multi-account Agentic AI Platform that platform teams operate, and workstream teams such as internal product and engineering teams consume through versioned, self-service interfaces.</p> -
<p>At enterprise scale, the developer experience plane is a catalog of organization-approved, versioned agent templates. The blueprint ships reference templates for <a href="https://github.com/strands-agents" target="_blank" rel="noopener">Strands</a>, <a href="https://www.langc…-
<p>Out of the four planes, two carry most of the architecture:</p> -
<ol type="1"> -
<li><strong>Agentic AI Platform control plane:</strong> The Platform account publishes versioned templates and interfaces, operates AWS Agent Registry governance records and shared Inference Gateways, admits models and Guardrails, orchestrates reviewed releases, and aggregates fle…-
<li><strong>Repeatable Workstream cell:</strong> Each product or domain receives an isolated boundary with a team-owned source repository, a Workload pipeline, and three Amazon Bedrock AgentCore capabilities: AgentCore Runtime, actor-scoped AgentCore Memory, and an AgentCore Gatew…-
</ol> -
<h3 id="key-components">Key components</h3> -
<p>The following diagram illustrates the operating model and governed flow of the solution.</p> +
<h2 id="event-overview">Event overview</h2> +
<p>Over three quarterly meetups this year, the UK community spent real time with Amazon Quick. Not slide decks about it, actual time building with it. People created <a href="https://docs.aws.amazon.com/quick/latest/userguide/custom-agents.html" target="_blank" rel="noopener">Amazon Quic…+
<h2 id="what-a-typical-event-looks-like">What a typical event looks like</h2> +
<p>All three events were held at the Amazon UK Corporate Office in Shoreditch, London. By 9:45 AM, the room is already buzzing. People reconnect over coffee, swap updates from the last quarter, and newcomers get pulled into conversations before the day even officially begins.</p> +
<p>From there, the energy carries through the full agenda: customers share real stories about what they’ve built with Amazon Quick, product leads walk through the latest updates, and panels spark the kind of back-and-forth you don’t get from a webinar. The 75-minute hands-on workshop is where …+
<p>The format has shifted each quarter. Q1 leaned more on introductions and demos because the group was meeting for the first time. By Q2, we added an icebreaker quiz, and ran a focused workshop on Quick Flows and automation. For Q3 we brought in a specialist panel, a group quiz, and a network…<div style="width: 810px" class="wp-caption alignnone">-
<a href="https://d2908q01vomqb2.cloudfront.net/f1f836cb4ea6efb2a0b1b99f41ad8b103eff4b59/2026/10/22/ML-21798-1.png" target="_blank" rel="noopener"><img src="https://d2908q01vomqb2.cloudfront.net/f1f836cb4ea6efb2a0b1b99f41ad8b103eff4b59/2026/10/22/ML-21798-1.png" alt="Operating model linking…-
<p class="wp-caption-text">Figure 1: Enterprise Agentic AI Platform operating model, showing the governance foundation, the Platform control plane, repeatable team-owned Workstream cells, and the software delivery, inference, tool, telemetry, and ownership flows between them.</p>+
<a href="https://d2908q01vomqb2.cloudfront.net/f1f836cb4ea6efb2a0b1b99f41ad8b103eff4b59/2026/10/28/ML-21295-1.jpeg" target="_blank" rel="noopener"><img src="https://d2908q01vomqb2.cloudfront.net/f1f836cb4ea6efb2a0b1b99f41ad8b103eff4b59/2026/10/28/ML-21295-1.jpeg" alt="Attendees at a UK Ama…+
<p class="wp-caption-text">Figure 1: Attendees at a UK Amazon Quick User Group meetup</p>+
</div> +
<h2 id="what-customers-shared">What customers shared</h2> +
<p>Customer stories have been part of the community from the start. Each speaker brought a different problem into the room, showing how teams are applying Quick in very different environments.</p> +
<p>At the first event, the conversation centered on getting insight into the hands of the people who need it. At <a href="https://www.arrivagroup.com/" target="_blank" rel="noopener">Arriva UK</a> Trains, critical data wasn’t reaching the people who needed it most. Train and statio…+
<p>By Q2, the discussion had moved toward a different challenge: getting the foundations right before introducing more AI and automation. At <a href="https://www.carsa.co.uk" target="_blank" rel="noopener">Carsa</a>, a used-car retailer, analytics functioned like a service desk. Qu…+
<p>Q3 brought the conversation further into automation and agentic workflows. <a href="https://en.propulse-lab.com/" target="_blank" rel="noopener">Propulse Lab</a>, a Swiss operating system for B2C subscription businesses, was handling 1 million support tickets a year. Their teams…+
<blockquote>+
<p><em>“What was once a cost centre became a growth engine – their product managers now own the entire system without writing a single line of code, and their support teams focus on proactive, revenue-driving conversations instead.”</em></p> +
<p>— Guirec Lecointre, Growth Operations Specialist, Propulse Lab</p>+
</blockquote> +
<p>The use cases vary widely, and that’s exactly the point. Rather than presenting a single vision for Amazon Quick, practitioners show what they’ve tried against real business problems, from governed analytics and frontline reporting to automation and AI. That shifts the conversation in the r…+
<h2 id="from-presentations-to-participation">From presentations to participation</h2> +
<p>Attendees changed across the three events. A listen-and-learn turned into building meaningful relationships by Q3. People were new to the group and new to each other. By Q2, the icebreaker quiz loosened things up early. The feedback sessions got more specific too: instead of “this was great…+
<p>Q3 pushes this further. Sofia Bourika, who joined the leadership team early on and presented the Arriva customer story at the first event, came up with an icebreaker called “What Are You Proud Of?”, attendees share something they’ve built with Amazon Quick. We’ve already got someone who was…+
<p>Our community stayed connected all year. Our LinkedIn group had weekly contributions and our email lists encouraged involvement.</p> +
<h2 id="collective-impact-in-action">Collective impact in action</h2> +
<p>A huge thank you to our dedicated AWS Partner, <a href="https://www.reply.com/data-reply/en" target="_blank" rel="noopener">Data Reply</a>, whose consistent contributions and thought leadership have been instrumental in making every UK user group meetup a success.</p> +
<p>Our UK leadership team led us to reach every milestone:</p> +
<p><strong>Liam Yusuph</strong>, AI and Data Delivery Lead, Data Reply</p> +
<p><strong>Priyanka Bettaswamygowda</strong>, Solutions Architect, AWS</p> +
<p><strong>Sahar Shahbaz,</strong> Data Analyst, Data Reply</p> +
<p><strong>Sofia Bourika,</strong> Lead BI Developer, Arriva</p> +
<p><strong>Tobi Daramola,</strong> Senior Data and Analytics Lead, Carsa</p> +
<p><strong>Ali Hayat</strong>, AV Support Associate, Amazon</p> +
<p><strong>Lam Phan</strong>, BI Developer, AWS</p> +
<p><strong>Rikhi Singh</strong>, ADC Engineer, AWS</p> +
<div style="width: 810px" class="wp-caption alignnone">+
<a href="https://d2908q01vomqb2.cloudfront.net/f1f836cb4ea6efb2a0b1b99f41ad8b103eff4b59/2026/10/28/ML-21295-2.jpeg" target="_blank" rel="noopener"><img src="https://d2908q01vomqb2.cloudfront.net/f1f836cb4ea6efb2a0b1b99f41ad8b103eff4b59/2026/10/28/ML-21295-2.jpeg" alt="The UK Amazon Quick U…+
<p class="wp-caption-text">Figure 2: The UK Amazon Quick User Group leadership team</p></div> -
<p>The architecture consists of the following account layers:</p> +
<h2 id="amazon-quick-community-resources">Amazon Quick Community resources</h2> +
<p>The <a href="https://community.amazonquicksight.com/" target="_blank" rel="noopener">Amazon Quick Community website</a> has tens of thousands of members and thousands of searchable questions. It also offers over 600 learning resources on accelerating your productivity with agent…+
<blockquote>+
<p><em>“The Quick Community is about learning together and building connections. It’s where users at any skill level can find help, learn something new, and meet others. What I love most about the Quick Community is how genuinely helpful everyone is. It’s a place that reveals possibilit…+
<p>— Kristin Mandia, Sr.&nbsp;Community Manager, AWS</p>+
</blockquote> +
<p>Learning opportunities include approximately two <a href="https://community.amazonquicksight.com/t/recurring-live-online-events/25696" target="_blank" rel="noopener">free live online sessions</a> per week, monthly feature update detailed and advanced automation sessions, and 4-h…+
<h2 id="key-takeaways">Key takeaways</h2> +
<p>A few themes came up consistently across the three events.</p> +
<h3 id="the-format-keeps-changing-because-we-listen">The format keeps changing because we listen</h3> +
<p>The Amazon Quick User Group UK runs on a set structure: an introduction, a product update, customer stories, a hands-on workshop, and a feedback session. But we’ve been adjusting it every quarter. Early on, we had the “What’s New with Quick” segment right after customer stories, and it brok…+
<h3 id="customers-present-not-us">Customers present, not us</h3> +
<p>At least one customer shares how they use Amazon Quick at each event. Across three events, we’ve had Carsa, Arriva, Propulse Lab, and a few other customers on stage. After the very first event, one of the clearest pieces of feedback was that people wanted more customer speakers. That told u…+
<h3 id="the-workshop-is-where-it-clicks">The workshop is where it clicks</h3> +
<p>Every event has a hands-on workshop, and people consistently tell us it’s the part they get the most out of. There’s a gap between watching someone demo a product and sitting down with it yourself. That gap is where most attendees say Amazon Quick starts to make sense for what they’re tryin…+
<h2 id="conclusion">Conclusion</h2> +
<p>The success of this year’s meetups is a testament to every single member who showed up, shared their knowledge, and contributed to making this community what it is today. As we look ahead, we’re more energized than ever to keep this momentum going and continue pushing the boundaries of what…+
<h2 id="get-involved">Get involved</h2> +
<p>To learn more about Amazon Quick and connect with the community:</p> +
<ul> +
<li>Join the UK Amazon Quick User Group on <a href="https://www.linkedin.com/groups/16948010/" target="_blank" rel="noopener">LinkedIn</a>.</li> +
<li>Join the <a href="https://community.amazonquick.com/" target="_blank" rel="noopener">Amazon Quick Community</a> to learn agentic AI together and access over 600 learning resources.</li> +
<li>Explore <a href="https://aws.amazon.com/bedrock/" target="_blank" rel="noopener">Amazon Bedrock</a> for additional generative AI capabilities.</li> +
<li>Register for the next UK Amazon Quick User Group meetup: <a href="https://luma.com/21k0l3fa" target="_blank" rel="noopener">January 28, 2027</a>.</li> +
<li>Not in the UK? Attend a meetup in <a href="https://community.amazonquick.com/t/user-groups-la-dach-chicago-phoenix-austin-nyc-boston-uki-to-come/47826" target="_blank" rel="noopener">Austin, Phoenix, Chicago, Boston, Los Angeles, New York City, or Munich/DACH</a>.</li> +
<li>Complete the <a href="https://pulse.amazon/survey/HIZTMBUZ?p=0" target="_blank" rel="noopener">leadership interest form</a> to start an Amazon Quick User Group in your city.</li> +
</ul> +
<p>For questions about the event or the Amazon Quick Community, Kristin Mandia, Sr.&nbsp;Community Manager, Amazon Quick Community and User Group program, at <a class="email" href="mailto:[email redacted]">[email redacted]</a>.</p> +
<p>&nbsp;</p> +
<p style="clear: both"> </p>+
<hr style="width: 100%"> +
<h2>About the authors</h2> +
<footer> +
<div class="blog-author-box" style="padding-top: 2.0em"> +
<div class="blog-author-image" style="margin-right: 1.0em">+
<img loading="lazy" class="alignnone size-full" src="https://d2908q01vomqb2.cloudfront.net/f1f836cb4ea6efb2a0b1b99f41ad8b103eff4b59/2026/10/28/ML-21295-4.jpg" alt="Priyanka Bettaswamygowda" width="100" height="133">+
</div> +
<h3 class="lb-h4">Priyanka Bettaswamygowda</h3> +
<p style="overflow: hidden">Priyanka is a Solutions Architect at AWS and has a great passion for Amazon Quick. Priyanka works with customers across the UK regional public sector to help adopt cloud technologies and accelerate innovation.</p> +
</div> +
<div class="blog-author-box" style="padding-top: 2.0em"> +
<div class="blog-author-image" style="margin-right: 1.0em">+
<img loading="lazy" class="alignnone size-full" src="https://d2908q01vomqb2.cloudfront.net/f1f836cb4ea6efb2a0b1b99f41ad8b103eff4b59/2026/10/28/ML-21295-3.jpg" alt="Liam Yusuph" width="100" height="133">+
</div> +
<h3 class="lb-h4">Liam Yusuph</h3> +
<p style="overflow: hidden">Liam is an AI &amp; Data Delivery Lead at Data Reply with a passion for AI-powered analytics and Amazon Quick. Liam leads the delivery of AI and data solutions, helping organizations turn emerging technologies into practical business value.</p> +
</div> +
</footer></content:encoded>+
+
+
+
</item>+
<item>+
<title>Building intent-driven UI testing with Amazon Nova Act</title>+
<link>https://aws.amazon.com/blogs/machine-learning/building-intent-driven-ui-testing-with-amazon-nova-act/</link>+
+
<dc:creator><![CDATA[Sakthi Chellapparimanam]]></dc:creator>+
<pubDate>Thu, 29 Oct 2026 18:20:27 +0000</pubDate>+
<category><![CDATA[Amazon Nova]]></category>+
<category><![CDATA[Customer Solutions]]></category>+
<guid isPermaLink="false">d49218181a0d92fa2228059f54218aae84fd0b63</guid>+
+
<description>Epsilon's Agentic Platform team delivers UI testing as a managed, multi-tenant service built on Amazon Nova Act and Amazon Bedrock AgentCore Browser Tool. Tests are written in plain English and run on browsers managed by AWS, giving every product team resilient, intent-driven testi…+
<content:encoded><p>To scale intent-driven UI testing across its product teams, <a href="https://www.epsilon.com/" target="_blank" rel="noopener">Epsilon</a>, a data, technology, and marketing-services company and part of <a href="https://www.publicisgroupe.com/" targe…+
<p>Within Epsilon, the Agentic Platform team (50 engineers) builds shared infrastructure so other product teams can build, deploy, govern, and run AI agents safely. With this shared platform, teams get a skills-based agent builder. They also get federated identity and pre-flight authorization,…+
<p>One of those managed capabilities is <em>UI testing</em>. With <a href="https://aws.amazon.com/nova/act/" target="_blank" rel="noopener">Amazon Nova Act</a> and <a href="https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/browser-tool.html" target="_blan…+
<p>In this post, we walk through Epsilon’s architecture for delivering UI testing as a managed platform service, show the core Nova Act and Amazon Bedrock AgentCore Browser Tool integration code, highlight a concrete 18-step test workflow, and share best practices and measured results.</p&g…+
<h2 id="the-challenge-enhancing-coverage-not-replacing-what-works">The challenge: Enhancing coverage, not replacing what works</h2> +
<p>Epsilon’s existing UI testing practice was not broken. The team maintained a <em>Playwright-based regression suite</em> that validated functionality across their web applications, and it worked well. Playwright remains their regression backbone today.</p> +
<p>The opportunity Epsilon identified was additive, not corrective:</p> +
<ul> +
<li><strong>Scripted Playwright tests</strong> are effective for deterministic regression checks. However, they are tightly coupled to selectors and page structure. Cosmetic UI changes can break them even when the underlying functionality is sound.</li> +
<li><strong>Multiple product teams</strong> needed full testing coverage against their own applications. Standing up browser infrastructure, managing dependencies, and integrating with identity systems per team created duplicative effort.</li> +
<li>The platform team wanted UI testing to be a <strong>managed capability</strong> that fit their existing agent model, not a custom CLI that every onboarded team had to operate.</li> +
</ul> +
<p>The goal: layer <strong>intent-based testing</strong> on top of the existing Playwright regression suite. Deliver it as a multi-tenant platform service so that each Epsilon product team could consume through the same API surface they used for every other agent capability.</p&…+
<h2 id="why-epsilon-chose-amazon-nova-act-and-the-amazon-bedrock-agentcore-browser-tool">Why Epsilon chose Amazon Nova Act and the Amazon Bedrock AgentCore Browser Tool</h2> +
<p>Epsilon evaluated options against two requirements: natural-language test authoring (to complement selector-based Playwright tests) and managed browser infrastructure (to fit their multi-tenant platform model). With Amazon Nova Act combined with Amazon Bedrock AgentCore Browser Tool, Epsilo…+
<ul> +
<li><strong>Natural-language steps.</strong> Test steps are plain-English instructions executed by Nova Act. They are resilient to cosmetic UI changes and readable by non-engineers. A step like <em>“Confirm the rewards balance is visible”</em> ages far better than a br…+
<li><strong>AWS managed, ephemeral browser sessions.</strong> With Amazon Bedrock AgentCore Browser Tool, teams get isolated browser instances with no binaries to ship and no display server to maintain. Amazon Bedrock AgentCore Browser Tool also offers live view for human observat…+
<li><strong>AWS Identity and Access Management (AWS IAM)-based authentication.</strong> With Nova Act, you can authenticate through the standard AWS credential chain (the task role). This fits Epsilon’s existing security model without distributing separate API keys.</li> +
<li><strong>Cross-account friendly.</strong> With Nova Act and Amazon Bedrock AgentCore Browser Tool, you can slot into your existing AWS Security Token Service (AWS STS) AssumeRole provisioning. Each onboarded team can run tests inside their own AWS account under a scoped role. T…+
</ul> +
<p>Epsilon chose Amazon Bedrock AgentCore Browser Tool specifically because it eliminates the operational burden of managing browser binaries, display servers, and session lifecycle. The Agentic Platform team had no interest in owning that infrastructure.</p> +
<h2 id="the-solution-architecture-and-implementation">The solution: Architecture and implementation</h2> +
<p>Epsilon’s solution has three main components: a platform API for creating workflows, a background task runner for executing them, and the Nova Act and Amazon Bedrock AgentCore Browser Tool integration that drives the browser.</p> +
<h3 id="platform-architecture">Platform architecture</h3> +
<p>Epsilon’s UI testing capability follows a clear pipeline:</p> <ol type="1"> -
<li><strong>Management and governance foundation.</strong> AWS Organizations, AWS IAM Identity Center, preventive policy, audit, retention, and cost policy establish the enterprise boundary. These are logical responsibilities that can map to an existing AWS Control Tower landing z…-
<li><strong>Security, audit, and observability layer.</strong> AWS CloudTrail, AWS Config, AWS Security Hub, Amazon GuardDuty, log archive, and an Amazon CloudWatch OAM sink provide independent control-plane evidence and fleet visibility. Central operators can query shared signals…-
<li><strong>Platform layer of Agentic AI Platform</strong>. Separate Platform environments own golden paths, the governance catalog, Guardrails, shared Inference Gateways, exact tool aliases, pipelines, compatibility, quotas, and common operational interfaces. Platform changes are…-
<li><strong>Workstream fleet layer.</strong> Each product, domain, or regulated boundary receives one or more nonproduction and production cells. Teams own Runtime, Memory, Tool Gateway, tools, data, prompts, evaluation cases, application incidents, and service-level objectives (S…+
<li>A user creates a workflow through the platform API, specifying a <code>target_url</code> and a list of natural-language steps.</li> +
<li>On create, the platform registers a Nova Act workflow definition on AWS.</li> +
<li>On execute, a background task runner (<a href="https://docs.celeryq.dev/" target="_blank" rel="noopener">Celery</a>, a Python framework for asynchronous job processing) opens an Amazon Bedrock AgentCore Browser Tool browser session. It runs each step through Nova Act over a &l…+
<li>The system captures a screenshot per step, tracks pass/fail per step, computes an overall pass rate, and stores everything (results JSON, screenshots in Amazon S3, session ID).</li> +
<li>A completion hook fires so the result is recorded back into the platform.</li> </ol> -
<h2 id="solution-walkthrough">Solution walkthrough</h2> -
<p>The following sections trace the key design decisions: account topology, preventive and detective controls, the AgentCore execution path, and the delivery pipeline.</p> -
<h3 id="account-topology-and-organizational-units">Account topology and organizational units</h3> -
<p>The blueprint defines logical account roles that separate enterprise governance, shared Platform services, and team-owned execution:</p> +
<div style="width: 810px" class="wp-caption alignnone">+
<a href="https://d2908q01vomqb2.cloudfront.net/f1f836cb4ea6efb2a0b1b99f41ad8b103eff4b59/2026/10/22/ML-21285-1.png" target="_blank" rel="noopener"><img src="https://d2908q01vomqb2.cloudfront.net/f1f836cb4ea6efb2a0b1b99f41ad8b103eff4b59/2026/10/22/ML-21285-1.png" alt="Epsilon UI Testing Serv…+
<p class="wp-caption-text">Figure 1: Epsilon’s UI Testing Service architecture</p>+
</div> +
<p>Figure 1 shows the layered architecture of Epsilon’s UI Testing Service. User requests flow from the Agent Cloud Frontend through an API Gateway with authentication and permission checks. The API layer (Django REST) exposes endpoints for workflows, executions, automation suites, and stats. …+
<p>The capability is exposed as a standard agent (<code>ui-test-agent</code>) with two skills:</p> +
<ul> +
<li><strong>ui-test-recorder</strong>: Explore a scenario interactively and record natural-language steps.</li> +
<li><strong>ui-test-executor</strong>: Run recorded steps and return structured results.</li> +
</ul> +
<p>Onboarded teams consume UI testing the same way they consume any other platform agent. No additional infrastructure to manage. The onboarding process is straightforward. The platform team provisions a cross-account IAM role for the new team, registers their target application URLs, and gran…+
<h3 id="the-core-integration">The core integration</h3> +
<p>The Nova Act + Amazon Bedrock AgentCore Browser Tool integration is concise:</p> +
<div class="hide-language"> +
<pre><code class="language-python">import boto3+
from bedrock_agentcore.tools.browser_client import browser_session+
from nova_act import NovaAct+
+
boto_session = boto3.Session(region_name=region)+
+
with browser_session(region) as client:+
ws_url, headers = client.generate_ws_headers()+
with NovaAct(+
cdp_endpoint_url=ws_url,+
cdp_headers=headers,+
boto_session=boto_session,+
starting_page=url+
) as nova_act:+
result = nova_act.act(instruction)</code></pre> +
</div> +
<p>This pattern (open a managed browser session, connect through the CDP WebSocket, and drive actions with natural language) is the foundation. Epsilon wraps it with their platform’s multi-tenant orchestration, result capture, and observability layers.</p> +
<h3 id="example-automating-the-create-an-agent-flow">Example: Automating the ‘Create an agent’ flow</h3> +
<p>To illustrate intent-based testing, here is a representative workflow from Epsilon’s platform. It exercises login, form entry, conditional handling, scrolling, and verification. The following steps highlight the differentiating patterns of Nova Act:</p> <table border="1px" width="100%" cellpadding="10px"> <tbody> <tr> -
<td><strong>Role</strong></td> -
<td><strong>Landing-zone placement</strong></td> -
<td><strong>Purpose</strong></td> +
<td><strong>Step</strong></td> +
<td><strong>Instruction</strong></td> </tr> <tr> -
<td>Management foundation</td> -
<td>Root / Security and Management</td> -
<td>Organizations, Identity Center, preventive policy, audit, retention, and cost policy</td> +
<td>1</td> +
<td>Type the username and choose <strong>Next</strong></td> </tr> <tr> -
<td>Security and log archive</td> -
<td>Security</td> -
<td>CloudTrail, AWS Config, Security Hub, GuardDuty, and retained audit data</td> +
<td>2</td> +
<td>Enter the password and choose <strong>Verify</strong></td> </tr> <tr> -
<td>Observability</td> -
<td>Security or Operations</td> -
<td>CloudWatch OAM sink and fleet logs, metrics, traces, and cost views</td> +
<td>3</td> +
<td>Navigate to <strong>Create Agent</strong> and select the <strong>Design a new specification</strong> card</td> </tr> <tr> -
<td>Agentic AI Platform</td> -
<td>Agent Platform</td> -
<td>Golden paths, AWS Agent Registry, Guardrails, Inference Gateways, pipelines, and fleet controls</td> +
<td>4</td> +
<td>If an onboarding tour box appears, choose <strong>Skip Tour</strong>, then verify the correct page loaded</td> </tr> <tr> -
<td>Workstream cell</td> -
<td>Agent Workstreams</td> -
<td>Team-owned Runtime, Memory, Tool Gateway, tools, data, evaluation, and application SLOs</td> +
<td>5</td> +
<td>Fill in the agent name and description fields</td> </tr> <tr> -
<td>Additional Workstream cells</td> -
<td>Agent Workstreams</td> -
<td>Repeat the same baseline per product, domain, environment, or regulated boundary</td> +
<td>6</td> +
<td>Scroll down until the <strong>Goal Statement</strong> field is visible, then fill it in</td> +
</tr> +
<tr> +
<td>7</td> +
<td>If the skills search box is not visible, scroll until it appears. Then search for and add a skill</td> +
</tr> +
<tr> +
<td>8</td> +
<td>Choose <strong>Save</strong>, confirm the success prompt, navigate back, and verify the new agent appears in the list</td> </tr> </tbody> </table> -
<p>This topology integrates into an existing landing zone, a pre-configured, secure, and scalable multi-account AWS environment, where the Management, Platform, and Workstream labels describe responsibilities rather than a fixed three-account estate. An organization can operate multiple valida…-
<h3 id="service-control-policies">Service control policies</h3> -
<p>Service control policies constrain the maximum permissions available to Platform and Workstream accounts. They complement, rather than replace, IAM, gateway authorization, Guardrails, resource policies, network controls, and pipeline evidence. The blueprint renders and reviews the effective…-
<ol type="1"> -
<li><strong>Model and Region admission:</strong> Organization policy and Gateway execution roles restrict inference to approved models and the supported Region. Adding a model or Region requires service availability, residency, IAM, quality, cost, rollback, observability, and deni…-
<li><strong>Guardrail and inference enforcement:</strong> Agents do not call model providers directly. Inference crosses the governed gateway, where the stage Guardrail and model allow-list are applied.</li> -
<li><strong>Pipeline-only Workstream mutation:</strong> Workstream teams open pull requests and use the Workload pipeline. Direct developer deployment into Workstream accounts is outside the supported operating model, which keeps the reviewed release and evidence path non-bypassab…-
<li><strong>Registry and Gateway governance:</strong> Workstream Gateway synthesis fails when a subscribed Registry record is missing, changed, or not approved.</li> -
<li><strong>Scoped deployment roles:</strong> Bootstrap policies are generated per account and Region, validated with AWS Identity and Access Management Access (IAM) Analyzer, and checked through effective-policy simulation. Administrator access is reserved for initial setup rathe…-
<li><strong>Separation of duties and adversarial proof:</strong> Platform controls model, Guardrail, Registry, and shared-service admission. Workstream teams control application behavior. Every denied path is paired with an authorized positive call and a mutation that proves the i…-
</ol> -
<p>An administrator inside a Workstream account can’t create a supported fast path around these controls. New or changed organization policies need a sandbox or representative-cell soak, exact allow and deny tests, and a rollback plan before broad attachment.</p> -
<h3 id="cicd-pipeline-with-evaluation-gates">CI/CD pipeline with evaluation gates</h3> -
<p>The blueprint uses reviewed Platform and Workload pipelines to enforce the following promotion sequence:</p> -
<ol type="1"> -
<li><strong>Source and review:</strong> A pull request establishes the immutable source revision. The pipeline pins the intended branch so an unmerged change can’t silently build the repository default branch.</li> -
<li><strong>Build and synth:</strong> The pipeline compiles code, runs tests and linting, creates the agent manifest, scans the digest-bound image, validates policies, resolves approved Registry descriptors, and inspects synthesized AWS CloudFormation and cdk-nag output.</li>…-
<li><strong>Stable roles and permission handoff:</strong> The Workload pipeline creates stable Workstream roles and pauses. The Platform pipeline grants exact tool aliases to the exact Gateway role, live policies are verified, and only then does the Workload pipeline continue.<…-
<li><strong>Deploy to nonproduction:</strong> The pipeline deploys Runtime, Memory, Tool Gateway, observability, alarms, budgets, tags, and application resources into the cell’s nonproduction boundary.</li> -
<li><strong>Evaluate the deployed agent:</strong> The gate invokes the real Runtime and measures regression, quality, tool success, refusal behavior, first-token latency, and cost. Every authorized call is paired with an adversarial twin: the gate calls an approved model and confi…-
<li><strong>Human approval:</strong> A reviewer examines the deployed evidence, known limitations, and residual risk. A generic green build is not enough when a required probe, resource, or expected denial is missing.</li> -
<li><strong>Promote, observe, and roll back:</strong> Production promotion uses the reviewed revision and continuity checks. Runtime sessions are sampled through update, cancellation, rollback, restore, and re-run so the pipeline proves both the forward path and recovery path.<…-
</ol> -
<p>No Workstream deployment step has a manual fast track. The Platform team owns the release contract, but Workstream teams operate their own pipeline instances and deployment cadence. Retirement follows a strict dependency order: Platform alias grants are revoked first, then Workstream Gatewa…-
<h2 id="security-and-governance">Security and governance</h2> -
<p>The blueprint implements defense in depth through reinforcing identity, policy, delivery, data, and evidence controls:</p> -
<ol type="1"> -
<li><strong>Identity and trust boundaries:</strong> AgentCore Identity, a capability of Amazon Bedrock AgentCore, and Amazon Cognito provide short-lived inference credentials. AWS_IAM and SigV4 authenticate tool traffic. Exact Runtime and Gateway roles, issuer and audience checks,…-
<li><strong>Model safety and admission:</strong> Inference flows through the governed gateway described in the Solution overview, applying Guardrails and model allow-lists before every call.</li> -
<li><strong>Tool authorization:</strong> The Tool Gateway invokes only Registry-approved aliases, as detailed in the Solution overview. Cedar policies provide authorization, with a retained AWS Lambda wrapper for defense in depth.</li> -
<li><strong>Tenant data protection:</strong> Workstream accounts isolate application state. AgentCore memory uses actor-scoped namespaces and customer-managed KMS keys. Tools and application data remain team-owned, and retention, deletion, and access policies are explicit rather t…-
<li><strong>Software supply chain and release:</strong> Reviewed source, manifest hashing, reproducible builds, digest-bound image scanning, policy validation, cdk-nag checks, deployed-runtime evaluation, human approval, rollback, and dependency-ordered teardown form one release c…-
<li><strong>Audit, observability, and cost governance:</strong> CloudTrail and retained control-plane data corroborate pipeline evidence. OAM shares logs, metrics, and traces with central operators. Standard allocation tags, budgets, quotas, and AWS Cost and Usage Report reconcili…-
</ol> -
<p>The repository is a reference implementation. Developers can adjust it to address their needs. The deployment and available features are based on the eu-west-1 (Ireland) Region. A substituted component, another Region, or a wider organizational rollout requires its own availability, securit…-
<h2 id="conclusion">Conclusion</h2> -
<p>In this post, we walked through an enterprise Agentic AI Platform built on Amazon Bedrock AgentCore. The architecture separates a centrally governed Platform control plane from repeatable, team-owned Workstream cells, with distinct inference and tool gateways, and production promotion gated…-
<p>The companion repository includes a blueprint that validates these controls end to end, from pipeline gates and adversarial boundary tests through zero-residual teardown. The support claim stops at that tested boundary.</p> -
<p>To get started, deploy one representative Workstream cell from the companion repository, prove its full lifecycle, and then automate additional cells from the same versioned baseline.</p> -
<p>Related resources:</p> -
<ol type="1"> -
<li><a href="https://github.com/aws-samples/sample-ai-agent-factory/tree/main/enterprise-agentic-ai-platform-blueprint" target="_blank" rel="noopener">Enterprise Agentic AI Platform Blueprint on GitHub</a></li> -
<li><a href="https://docs.aws.amazon.com/prescriptive-guidance/latest/agent-ops-maturity-roadmap-for-enterprise-ai/introduction.html" target="_blank" rel="noopener">Enterprise multi-account architecture for running agentic AI in production on AWS (Prescriptive Guidance)</a></li…-
<li><a href="https://catalog.us-east-1.prod.workshops.aws/workshops/3f49be39-c62b-40a2-975b-be9bf626526a/en-US" target="_blank" rel="noopener">Building an Enterprise Agentic AI Platform on Amazon Bedrock AgentCore Workshop</a></li> -
<li><a href="https://aws.amazon.com/bedrock/agentcore/" target="_blank" rel="noopener">Amazon Bedrock AgentCore documentation</a></li> -
</ol> +
<p>Steps 4 and 7 demonstrate <em>conditional logic</em>: “if an element appears, handle it” and “if a target isn’t visible, scroll until it is.” These patterns would require explicit branching in a traditional framework. Nova Act handles them naturally through language understandin…+
<h3 id="runtime-isolation">Runtime isolation</h3> +
<p>Epsilon runs UI tests in a <em>dedicated runner service</em> (a FastAPI sidecar) that pins <code>nova-act</code> 3.x with Workflow and IAM authentication support, plus the Amazon Bedrock AgentCore Browser Tool client. This keeps dependencies separate from the main ba…+
<h2 id="best-practices">Best practices</h2> +
<p>Based on their experience building UI testing as a platform capability, the Epsilon team recommends:</p> +
<h3 id="write-steps-as-intent-not-mechanics">1. Write steps as intent, not mechanics</h3> +
<div class="hide-language"> +
<pre><code class="language-plaintext">GOOD: "Confirm the rewards balance is visible"+
BAD: await page.locator('#rewards-balance-container &gt; span.amount').toBeVisible()</code></pre> +
</div> +
<p>Intent-based steps survive UI redesigns. Product managers and quality assurance (QA) leads who might not write code can review them directly.</p> +
<h3 id="isolate-the-nova-act-runtime">2. Isolate the Nova Act runtime</h3> +
<p>Pin Nova Act and Amazon Bedrock AgentCore Browser Tool dependencies in a dedicated service or container. Epsilon learned that the signing endpoint for Nova Act targets <code>us-east-1</code> regardless of the browser session’s region. Running it in a separate service with its ow…+
<h3 id="reuse-recorded-trajectories-to-cut-cost">3. Reuse recorded trajectories to cut cost</h3> +
<p>Epsilon uses trajectory retries in <strong>draft state</strong>, saves the successful trajectory, and replays it in <strong>active state</strong>. Repeat runs skip redundant large language model (LLM) “thinking,” reducing both latency and cost for regression-style re…+
<h3 id="design-for-multi-tenancy-from-day-one">4. Design for multi-tenancy from day one</h3> +
<p>Use cross-account AssumeRole with scoped IAM policies. Each tenant’s tests run under their own role in their own account. This provides isolation, auditability, and failure scope containment without custom infrastructure per team.</p> +
<h3 id="lean-on-managed-observability">5. Lean on managed observability</h3> +
<p>With Amazon Bedrock AgentCore Browser Tool, teams get live view, session recording, and AWS CloudTrail integration. Epsilon uses these built-in capabilities instead of building custom observability, reducing maintenance and giving compliance teams familiar audit surfaces.</p> +
<h3 id="account-for-amazon-bedrock-agentcore-browser-tool-viewport-behavior">6. Account for Amazon Bedrock AgentCore Browser Tool viewport behavior</h3> +
<p>When prompting Nova Act steps, account for the browser viewport defaults of Amazon Bedrock AgentCore Browser Tool. Steps that reference scrolling (for example, “scroll down until the Goal Statement textbox is visible”) verify that Nova Act handles content in the lower portion of the page co…+
<h2 id="results-and-impact">Results and impact</h2> +
<p>Epsilon’s UI testing capability is operational across multiple teams, with the multi-tenant architecture proven in production.</p> +
<table border="1px" width="100%" cellpadding="10px"> +
<tbody> +
<tr> +
<td><strong>Dimension</strong></td> +
<td><strong>Amazon Nova Act</strong></td> +
<td><strong>Playwright (equivalent)</strong></td> +
</tr> +
<tr> +
<td>Lines authored</td> Diff display stops at 400 lines. The line counts above are from the whole diff. 78 lines shown here cut at 300 characters. 2 email addresss masked in this display. See the About page on personal data. The raw artifact at this commit is linked above.