Aquarela SmartCode

AI-assisted development, with engineering rigor

We bring Kiro, the AWS agentic AI development environment, to your team: licenses, onboarding, project standards, and usage and cost governance. As an add-on, AI code review on pull requests with Amazon Bedrock.

Talk to a specialist See how it works

Kiro

AI agent IDE and CLI, from AWS

Spec-driven

requirements, design and tasks before code

Governance

SSO, usage and cost control per developer

The challenge

AI in the editor speeds up typing. It does not always speed up delivery.

In lean teams, every bit of rework hurts. When AI goes straight from prompt to code, intent gets lost in the chat history and problems show up late, in review or in production.

Implicit requirements

Whatever the prompt left out becomes a bug. Edge cases only surface after the code is written.

Standards live in senior heads

Team architecture and conventions are passed on case by case, in every pull request review.

Forgotten glue work

Tests, docs and fixtures go stale between sprints because they depend on someone remembering.

Unpredictable cost

Scattered subscriptions, no view of who uses what, how much is consumed or the monthly spend cap.

How it works

Specification first, code second.

Kiro turns a natural language request into requirements, design and tasks. The team reviews and approves the spec before any implementation.

Specify

Describe

The goal is written in natural language, without a single line of code.

Requirements

Kiro generates requirements.md with user stories and acceptance criteria.

Design and tasks

design.md captures architecture and technical decisions; tasks.md, the implementation sequence.

Build

Human review

The team approves or adjusts the spec before implementation.

Implementation

The agent runs the tasks, writes the files, runs tests and iterates.

Reviewed pull request

The approved spec travels with the PR and gives reviewers an objective reference.

Maintain

Hooks

Automatic actions triggered by events, such as running tests or updating docs.

Steering

Team standards and conventions are part of the context of every agent task.

The gain is not typing faster: it is cutting the rework caused by implicit requirements.

Kiro features

What your team uses day to day.

Specs

Requirements, design and sequenced tasks, versioned alongside the code like any other artifact.

Steering files

Markdown files that give Kiro persistent knowledge of the project, its architecture and conventions.

Agent Hooks

Commands or prompts that run automatically when events happen in the session, such as saving a file.

MCP and Powers

Connects the agent to the tools and systems your team already uses, through open standards such as MCP.

Checkpoints

Every interaction creates a restore point: undo the agent's changes and return to an earlier state.

Kiro CLI

The same agent in the terminal and in pipelines, for command line workflows and automation.

What Aquarela delivers

From license to productive team, with governance from day one.

We do not just hand over tool access. We set up the way of working with it, from pilot to scale, with usage and cost control.

Kiro licenses

Sizing and purchase of the right plans for each developer profile, with centralized seat management.

Onboarding and workshop

Hands-on spec-driven development workshop using a real feature from your backlog, from request to pull request.

Team standards in the agent

Steering files with documented architecture and conventions, plus hooks for tests and docs.

Identity and access

SSO through AWS IAM Identity Center, individual access per seat and least-privilege permissions.

Usage and cost control

Credit consumption tracked per developer, a defined overage policy and a predictable spend cap.

AI code review

Add-on

Automated pull request review with Amazon Bedrock models, with feedback in the developer workflow and support for vulnerability detection.

Kiro plans

Per-developer subscription, with monthly credits.

Each plan includes a monthly credit pool, the unit of work the agent consumes. We help pick the right plan for each profile based on real usage measured in the pilot.

Plan	Credits per month	Usage profile
Kiro Free	50	Assess whether the spec-driven flow fits the team.
Kiro Pro	1,000	Occasional agent use throughout the week.
Kiro Pro+	2,000	Kiro as a regular part of the workday.
Kiro Pro Max	5,000	Intensive development, several hours a day.
Kiro Power	10,000	Continuous, high-volume use.

Team plans include consolidated billing, usage analytics and SSO through AWS IAM Identity Center.
Plan credits do not roll over to the next month; add-on credits can be purchased on paid plans.
On paid plans, your content is not used to train models and generated output is covered by IP indemnity, under AWS terms.

Check current prices on the official Kiro pricing page

Phased adoption

Start small, measure, then scale.

Built for lean teams: the first value shows up in the pilot, on a real feature, and the scale decision is made with usage data.

PHASE 1

Pilot on one repository

One real feature through the full cycle
Credit consumption measured

PHASE 2

Standards and steering

Team conventions documented
Spec review checklist

PHASE 3

Automation with hooks

Hooks for tests and docs
Integrated into the pull request flow

PHASE 4

Scale with governance

Seats and SSO via IAM Identity Center
Overage policy and cost dashboard

AWS services

Built on AWS.

SmartCode uses native AWS tools and services, with identity, access and cost under the governance your company already knows.

Does your team use Amazon Q Developer in the IDE?

AWS ends support for the Amazon Q Developer IDE plugins on April 30, 2027 and recommends Kiro as the replacement. We plan the transition with a pilot, standards and governance.

Read the AWS notice