Complete AI Training
Start for Free

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

AI news ·

AWS Quick adds flows you build in chat, multi-agent orchestration, and an updated web experience

AWS Quick now lets teams turn one person's workflow into a shared, automated process through conversation, with one company saving 80 hours per month from a single use case.

AWS Quick released an update that lets teams turn one person's workflow into a shared, automated process through conversation, run agents that work together on a single request, and manage work from a redesigned web app that keeps tasks running even after the browser closes. The features are available now in all supported AWS Regions.

The release addresses a common friction point: individual productivity gains from AI rarely spread across a team. The way one analyst builds a monthly report seldom becomes the team's standard, and work stalls when someone steps away. Quick's new capabilities target that gap by making processes repeatable, shareable, and persistent.

Building flows by describing them in chat

Quick Flows now lets users build, run, and refine automated workflows entirely through conversation. A user describes a process in chat the way they would explain it to a colleague, and Quick constructs a flow that orchestrates agents, skills, tools, and data. A diagram appears beside the chat, showing each step so the user can confirm the structure matches their intent.

These flows handle longer-running processes with larger data volumes. Deterministic steps - fixed calculations and checks - keep results consistent, while agents handle drafting and analysis. The human decides what goes into the final output. For example, a financial planning manager could describe a monthly variance report, and Quick would build a flow that pulls results from a dashboard, compares figures to a budget spreadsheet using set formulas, flags regions that moved beyond thresholds, drafts commentary, and posts a draft to the team's shared space and Slack.

Once published, anyone on the team can run the flow. The process produces the same structure each month, regardless of who initiates it.

Rodrigo Marini, Head of Data at Pomelo, a Latin American fintech provider, described the broader effect: "At Pomelo, AWS Quick changed who gets to build solutions. Before, every analysis or automation depended on the data team. Today, people across every area explore information, make decisions, and build their own agents and flows within a unified, governed environment." He added that nearly four times as many people now work with data each month without expanding the team, and a single use case saves 80 hours per month.

A redesigned web experience that keeps work moving

The updated Quick web app surfaces items needing attention on the home screen immediately after sign-in. A feed and a daily briefing consolidate emails and messages in priority order. Users can ask Quick about any item or have it draft replies for review.

Agents continue working after the browser tab closes. A user can schedule a task or start a longer request before a meeting, close the browser, and return later to review results in Mission Control. The Library section brings together recent work, favorites, pinned items, and files created in Quick. Conversations, agents, and files sync across web, desktop, and mobile.

Multiple specialized agents on one request

Multi-agent orchestration lets the agent a user asks pull in several specialized agents simultaneously, each working on the part it knows best. The initiating agent combines their outputs into a single response. Each specialized agent keeps its own instructions, knowledge, and context, so domain experts can build and maintain them independently.

For instance, asking why shipping costs rose in a region could trigger a forecasting agent built on planning data and a procurement agent built on vendor contracts. The combined response shows what each agent contributed.

Connectors to Meta ads and Amplitude

New connectors let Quick pull data from Meta ads and Amplitude. Users can create, manage, and analyze Meta ad campaigns in natural language through Meta's ads Model Context Protocol server. Quick combines that ad data with first-party and third-party sources and asks permission before taking any write action, such as creating a campaign or changing a budget. The Amplitude connector lets product and marketing teams query product usage, customer behavior, and conversion data in natural language.

Skills on web, guardrails, and more languages

Skills - reusable instructions that teach Quick to do a task a specific way - are now available in the web app. Users can browse, install, create, and manage them as they can on desktop.

Administrators gain several new controls. Configurable content safety and sensitive-information policies can filter prompts, block denied topics, detect prompt attacks, and protect personal information. These policies can run in detect-only mode before enforcement and apply to individuals, identity provider groups, or the whole account. Administrators can also assign custom permissions and usage limits by group, and control which data source types people can create, update, or share. Quick now supports Japanese, German, Chinese (Simplified and Traditional), and Vietnamese.

Why this matters for executives and operations teams

For leaders in operations, finance, marketing, and strategy, this release reduces the risk that critical processes live in one person's head. Standardizing a monthly report, a campaign analysis, or a budget review as a reusable flow means the work gets done the same way during vacations, turnover, or busy periods. The multi-agent capability means specialized knowledge from different departments - procurement, forecasting, legal - can combine in one response without forcing anyone to build a monolithic agent. And the persistent web experience means agents finish work while you are in meetings, not sitting idle waiting for you to return to your desk.

Share