AWS Frontier Agents saddle up to extend software development teams

Developers, developers, developers… right?

Wrong, it’s now developers, developers, agentic service constructs designed to provide autonomous app design, code reviews and penetration testing that learn over time while working independently.

AWS wants to extend bouncing-Ballmer’s initial call to arms with new frontier agents that form an invaluable part of modern software application development teams. AWS says that frontier agents are a new, more sophisticated class of AI agents with three defining characteristics.

Three agentic characteristics

First, they’re autonomous i.e. when a developer directs them toward a goal, the agents figure out how to achieve it.

Second, they’re scalable i.e. they can perform multiple tasks at the same time and distribute work across multiple agents.

Third, they work independently i.e. they can operate for hours or days without requiring intervention.

  • Kiro autonomous agent is a virtual developer that maintains context and learns over time.
  • AWS Security Agent is a virtual security engineer that helps build secure applications by being a security consultant for app design, reviews and testing.
  • AWS DevOps Agent is a virtual operations team member that helps resolve and proactively prevent incidents, while continuously improving an application’s reliability and performance.

The company says that coding tools have accelerated individual tasks, but many of them have also introduced new friction. Using these tools, AWS says that developers often to find themselves acting as the “human thread” that holds work together i.e. rebuilding context when switching tasks, manually coordinating cross-repository changes and restitching information scattered across tickets, pull requests and chat threads.

Breaking the human thread

So then, what would it take for a tool to cut this friction, so teams can stay focused and ship code faster? AWS claims that the Kiro autonomous agent keeps work moving independently while developers focus on priority tasks.

Kiro autonomous agent maintains persistent context across sessions and continuously learns pull requests and feedback. It can handle a range of tasks – from triaging bugs to improving code coverage with a single change spanning multiple repositories. Developers can ask it questions, describe a task and assign tasks in backlog directly from GitHub.

The agent will then independently figure out how to get the work done, sharing changes as proposed edits and pull requests, so developers stay in control of what gets incorporated.

“For teams, the Kiro autonomous agent is a shared resource that works alongside the entire team, building a collective understanding of a codebase, products and standards. It connects to a team’s repos, pipelines and tools, like Jira, GitHub and Slack, to maintain context as work progresses, adapting to changes or updates. Every code review, ticket and architectural decision informs the agent’s understanding, making it even more useful for the team over time,” detailed AWS, in a technical statement.

Security teams face a challenge as they need to proactively identify risks in development, while also reacting quickly when issues emerge.

AWS says that current tools often provide generic recommendations and penetration testing takes so much time and resources that it can’t keep up with fast-moving development teams. AWS Security Agent helps build secure applications across AWS, multicloud and hybrid environments.

The agent embeds deep security expertise throughout the development lifecycle, proactively reviewing design documents and scanning pull requests against organisational security requirements and common vulnerabilities. Users define an organisation’s security standards once and AWS Security Agent automatically validates them across applications during its review – helping teams address the risks that matter to their business, not generic checklists.

The agent also transforms penetration testing from a manual process into an on-demand capability and teams can expand penetration testing across an entire application portfolio. The agent returns validated findings with remediation code to fix the issues it finds. By continually validating security from design to deployment, the agent helps prevent vulnerabilities early.

Modern distributed applications – with microservices, cloud dependencies and telemetry spread across multiple tools – make it difficult to isolate issues and understand system behaviour. At the same time, as services scale, operations can continue to eat up more of time.

Application resources & telemetry mapped

AWS DevOps Agent delivers “fewer alerts and more sleep” for a team, says the company. It does this through always-on incident triage, guided resolution and recommendations for how to continuously improve the reliability and performance of applications across AWS, multi-cloud and hybrid environments.

“AWS DevOps Agent is on call when incidents happen, instantly responding to issues and using its knowledge of an application and the relationship between components to find the root cause of the problem. It learns resources and their relationships spanning everything from observability tools, like Amazon CloudWatch, Dynatrace, Datadog, New Relic and Splunk, to runbooks, code repositories and continuous integration and continuous delivery (CI/CD) pipelines. It maps application resources and correlates telemetry, code and deployment data to precisely pinpoint root causes and reduce mean time to resolution,” noted AWS.

So it’s true, it’s no longer developers, developers, developers, it’s developers, developers, and agentic software application development services for autonomous design, DevOps and independent testing at scale. Do you think Ballmer could have got all that out in one gasp without passing out?