Amazon Quick just got a massive upgrade at AWS Summit New York 2026 🚀
On June 17, 2026, AWS took the stage in New York and made it crystal clear that the conversation around artificial intelligence in the enterprise world has leveled up. It’s no longer about testing AI in pilot projects or debating what it might do someday. It’s about putting agents to work for real, right now, solving actual problems inside companies with measurable impact and without that nagging feeling you’re betting on something experimental.
And what really stood out at the event was exactly that: the announcements didn’t come loaded with vague promises. They came with ready-to-go products, concrete use cases, and a very clear direction for where AWS wants to take its customers. Among the day’s highlights were AWS Continuum, a native AI security service that operates at machine speed, and AWS Context, a context layer that finally gives agents the map they needed to navigate enterprise data. But it doesn’t stop there. Amazon Quick gained autonomous agents, Kiro landed on iOS, and there was plenty of new stuff with AWS DevOps Agent, AWS Transform, and Amazon Bedrock AgentCore. There’s a lot to unpack, so let’s break it all down 👇
AWS Continuum: agentic security at machine speed
When AWS introduced AWS Continuum for code vulnerabilities on the Summit stage, it became clear the company isn’t just slapping AI on top of existing services. It’s rethinking how security works when threats move in milliseconds and humans simply can’t keep up on their own. Continuum is a security service built natively to operate with AI agents, covering the full cycle of code vulnerability management. It continuously discovers vulnerabilities, validates which ones are genuinely exploitable, prioritizes them by business context, and helps remediate them across the entire stack, all within guardrails you define. This isn’t just a technical upgrade — it’s a paradigm shift in how companies will think about data protection and infrastructure going forward.
What makes Continuum especially interesting is that it was designed to be model-agnostic, leveraging the strengths of different AI models where each one excels and integrating new models as they emerge. AWS even mentioned the rise of security-specialized models like Claude Mythos as validation of this approach. At every step of the process, you have full visibility into what Continuum is doing, why it suggested a specific action, and what would happen if that action were rolled back. Every decision is explainable, every action is auditable, and every outcome feeds back into the system to improve the next cycle. Security teams stay in control of strategic decisions, but the repetitive, high-speed operational tasks go to the agents.
Beyond the core service for code vulnerabilities, AWS also launched Continuum threat modeling, which automatically generates comprehensive threat models from design documents or source code, delivering results in industry-standard format. For companies that need to scale security without necessarily tripling the size of their team, this kind of intelligent automation makes all the difference. And since the service is AWS-native, it integrates naturally with the entire infrastructure customers already use — no complex implementations or steep learning curves required.
AWS Context: the map agents needed to navigate your data
AWS Context solves a problem that anyone who has worked with AI agents inside companies knows all too well: the agent doesn’t know where the data is. It sounds simple, but it’s one of the biggest bottlenecks in enterprise AI adoption. Context is a new service that automatically builds a knowledge graph from a company’s existing data, inferring relationships between data assets, business rules, and domain knowledge. All of that becomes available to every agent across the organization, helping them arrive at the right answer much faster.
The logic behind Context is powerful: context is what makes an agent’s tenth decision better than its first. With the right context, your agent can see the latest interactions you had with a customer in the CRM and recommend the best follow-up. Without context, the agent is more likely to deliver recommendations that sound confident but are flat-out wrong. Companies have access to all kinds of data — from database records to Slack messages, documents, and emails. For an agent to make that information useful, it needs to understand which tables exist, what’s stored in different columns, which sources are most reliable, and how they all relate to each other.
Receive the best innovation content in your email.
All the news, tips, trends, and resources you're looking for, delivered to your inbox.
By subscribing to the newsletter, you agree to receive communications from Método Viral. We are committed to always protecting and respecting your privacy.
Built on the same knowledge graph technology that powers Amazon Quick, AWS Context works as an agentic search layer for your organization’s data, pluggable into all your agents. It comes with built-in governance to ensure agents only access the information they’re authorized to see. With all your data source metadata stored in Iceberg format on S3 Tables, you can build against AWS Context using the tools you already have. No infrastructure to provision. No retrieval pipeline to build.
And here’s the part that really scales: as agents interact with AWS Context, it learns which sources produce correct results, which paths are most commonly used, and which business rules matter — getting better over time. Each agent can then improve based on the findings from a single query. For example, a customer support agent triaging an order issue might need to check purchase history, shipping status, and return eligibility across multiple different sources. The next time an agent faces a similar problem, it already knows exactly where to go, cutting down search time and reaching a faster resolution.
Amazon Quick with autonomous agents: what this actually means in practice
Amazon Quick is the AI assistant built for the way you actually work, with enterprise security your company will genuinely approve. AWS made a point of differentiating its approach from other vendors: while some offer AI assistants locked inside a single application — creating what AWS calls walled gardens — others offer assistants with little concern for security and data sharing, the so-called wild gardens. With Quick, you don’t have to make that trade-off.
The big news from the AWS Summit is the arrival of autonomous agents in Quick. You can now create agents that work in the background with specific expertise, a defined communication tone, and access to tools. You could build a finance agent that processes orders as they come in, or a sales agent monitoring CRM interactions, emails, and Slack to proactively draft follow-ups, flag risks, or recommend next steps. And the best part: no coding required — Quick’s agents are accessible to anyone in the company.
Alongside autonomous agents, AWS also launched a new activity feed personalized to the way each person works. It consolidates email, messages, calendar, and tasks into a single prioritized view, and learns which messages you always respond to quickly, which threads you skip, and which topics drive your week. On top of that, while Quick already connected to the most popular apps, AWS announced 16 new native integrations with companies like Adobe, Moody’s, and Snowflake, further expanding the assistant’s ability to operate effectively in day-to-day workflows.
It’s also worth highlighting that the integration between Amazon Quick and AWS Context is one of the most powerful aspects of this combination. Since Context already delivers agents the map of the company’s data environment, Quick can operate with much greater precision and autonomy. The agent knows exactly which datasets are available, which sources are reliable for each type of analysis, and what permissions it has to access sensitive information. This synergy between the two services is a clear example of how AWS is building an ecosystem where the pieces fit together intelligently — not just coexist in the same product catalog.
Kiro on iOS: orchestrate development agents from your phone
The moment companies are experiencing around AI-assisted software development is nothing short of remarkable. AWS shared an example that illustrates this acceleration perfectly: Dhan, a unicorn fintech startup from India, used Kiro to build a new charting platform with a single engineer in just 8 weeks, when the original estimate was 12 to 24 months with a dozen people. Having a large dev team is no longer the bottleneck for the next big idea — but having access to the right tools anywhere you go can be.
That’s why AWS announced that Kiro is now available on iOS devices. You can now start a new project, track progress, direct an agent, or interact with your Kiro session right from your phone. With Kiro on mobile, agents no longer live only on your laptop. They run in an always-on cloud session that works whether you’re at your desk or not. You can kick off feature development on the subway, review code between meetings, or approve changes over lunch. When you get back to your laptop, you’ll be exactly where the agent left off. Same session, same context, ready to keep going. And while access has never been easier, everything still runs securely in your cloud environment. Full mobility, zero compromises.
AWS DevOps Agent with Release Management: agents that help you ship code safely
Writing code faster than ever with agents like Kiro is great, but you only start benefiting from that code when it hits production. Your coding agents can make you ten times more productive, only for your pipeline to not have adapted to the new pace. Pull requests waiting. Tests running sequentially. Failures popping up in production. Fixing this means shifting tasks, testing, and troubleshooting earlier in the cycle.
AWS expanded the AWS DevOps Agent with a new release management capability so you can ship code faster and more safely. The DevOps Agent now supports everything from release readiness reviews to testing, making sure your pipeline can keep up with development velocity.
Here’s how it works in practice: when you’re developing with Kiro or Claude Code, the moment you generate code, you can trigger release readiness reviews right from where you’re coding. For example, imagine a developer renames a parameter to make it clearer. The change looks small in isolation and local tests say everything’s fine, but the update could cause a break when interacting with the rest of the application. Because the developer is using the DevOps Agent, they discover the code’s impact before it goes to production — getting a summary of the issue and a recommended fix. Once the code is ready, you open a pull request and the DevOps Agent generates and runs test plans specific to the changes, catching regressions, UX issues, and integration failures before they reach production.
AWS Transform: continuous modernization to keep technical debt in check
When you start moving code to production faster than ever, a new challenge shows up. How do you keep all that software up to date? The truth is, the moment you deploy code, it starts aging. Dependencies go stale. Frameworks get deprecated. Before you know it, technical debt has piled up.
AWS Transform is AWS’s agentic AI service designed to support modernizations at scale. It has already eliminated more than 1.6 million hours of manual effort for customers like BMW Group, Experian, and others. Until now, Transform was something you ran as a one-off, pointing it at a specific problem. But with the speed agents are writing new code, you need an agent that continuously keeps your code updated and well-documented.
That’s why AWS launched AWS Transform with continuous modernization — an autonomous, always-on capability for software portfolio management and modernization. While your agents write new code, AWS Transform works behind them, finding technical debt, fixing it, validating the fix, and learning from each transformation to make the next one better. It connects to your existing pipeline tools like CodePipeline, Jenkins, GitHub Actions, and GitLab, fitting into your workflows seamlessly. With this capability, every software package stays current in a continuous cycle, so technical debt never gets the chance to become a risk.
Amazon Bedrock AgentCore: build, connect, and optimize agents in production
AWS is building for a future where billions of agents are in operation. And as customers move in that direction, one thing became clear: building an agent is easy, but getting it to production can be complex. That’s why the company built Bedrock AgentCore — the platform for when you want to move agents from proof of concept to production.
The approach is clearly resonating with the market. Over the past 6 months, the number of tasks executed by agents on AgentCore grew 15x. The PGA Tour is writing tournament coverage 10 times faster. And Nasdaq, Visa, and Experian are scaling agents across all their operations. At the AWS Summit, the company announced several improvements to make AgentCore even more comprehensive:
- Amazon Bedrock Managed Knowledge Base: a fully managed knowledge base that handles ingestion, parsing, and retrieval for your RAG and knowledge bases. It comes with native connectors for popular sources like S3, SharePoint, Confluence, and Google Drive, plus an agentic retriever for complex queries. Managed Knowledge Base integrates easily with AWS Context, enabling agentic search across all structured, unstructured, and domain data.
- Web Search in AgentCore: agents can now access up-to-date, accurate information from the web using the same search that already powers Quick, Kiro, and Alexa+. As a fully managed tool, Web Search returns re-ranked, current results operating natively within the AWS environment, ensuring your data and queries never leave AWS.
- New optimization capabilities: turn production traces into continuous improvements. You can now see failure, intent, and trajectory insights across hundreds of agent sessions, helping you understand what your agents are doing and how to make them better. AWS also announced general availability of recommendations and A/B testing to help test changes and improve agent performance.
- New security policy integrations: Amazon Bedrock Guardrails is now integrated with AgentCore, letting you evaluate every agent action against prompt injection attempts, harmful content, and sensitive data exposure. Soon, customers will also be able to feed in detection signals from leading security providers like Check Point, Zscaler, Rubrik, Netskope, and SentinelOne.
- AgentCore Harness now available: go from idea to working agent in minutes. With AgentCore Harness, you simply declare what your agent does, which model it uses, which tools it calls, and which instructions it follows — and AgentCore takes care of the rest, assembling the orchestration loop, tool execution, memory management, context, and error recovery.
Southwest Airlines accelerates AI adoption with AWS
The AWS Summit New York 2026 also featured a heavyweight partnership. Southwest Airlines announced it has chosen AWS as its preferred cloud provider to modernize its technology foundation and apply AI to transform how the airline operates and serves its customers. As part of the partnership, Southwest will transition from a predominantly on-premises environment to an AI- and agent-enabled cloud-based architecture on AWS by 2028, building a foundation for greater speed, flexibility, and reliability across the entire business.
More than 2,700 Southwest developers are already using Kiro to build features, automate testing, and generate cloud infrastructure to modernize Southwest.com. The airline is also adopting an AI-driven development lifecycle approach, where agents help move development forward while engineering teams guide and validate the results. As Southwest continues modernizing its presence on AWS, it’s expanding AI- and agent-based capabilities across the business, adopting new tools like Amazon Quick.
The compound effect of AI agents
One of the most important points raised during the AWS Summit is the concept of compound momentum that agents create. The idea is simple but powerful: the more you use agents, the more they deliver. More interactions give agents more context. More context leads to better results. Better results increase the trust you place in them. More trust means more work delegated. This virtuous cycle widens the gap between companies that genuinely adopt AI and those that just sit on the sidelines watching.
And that’s exactly what the Summit launches demonstrate in practice. AWS Continuum accelerates security. AWS Context gives agents the knowledge they need. Amazon Quick turns data into action. Kiro on mobile frees development from the desktop. The DevOps Agent ensures code reaches production with quality. AWS Transform keeps everything up to date. And Bedrock AgentCore orchestrates all of it at scale. Each piece feeds the next, creating an ecosystem where value multiplies as more agents come online.
What became clear at this AWS Summit New York 2026 is that AWS isn’t just offering standalone AI services. It’s building a cohesive platform where each piece amplifies the others — and the result is an ecosystem that’s increasingly hard to ignore for anyone taking artificial intelligence seriously in the enterprise. The last six months have already shown a seismic shift in how companies approach agents, moving from talk to action. The coming months promise to accelerate that even further. 🧠
