Natera optimizes lab operations with SWARM agentic AI

Executive Briefing
“Natera's lab engineering team partnered with Amazon Web Services to build SWARM for...”
Key Takeaways
- 1.Natera's SWARM system delivers about 130x faster first response time and 96.2 percent automated coverage across support tickets.
- 2.The solution runs on Amazon Bedrock AgentCore, integrating tools like Jira and Confluence alongside clinical regulatory security standards.
- 3.The team designed a skill library of 33 components and split tasks between deterministic Python code and large language models.
Natera's lab engineering team partnered with Amazon Web Services to build SWARM for automating equipment troubleshooting workflows. The system drastically reduces initial response times and automates the vast majority of support tickets.
Key Developments and Technical Context
Natera's SWARM system delivers about 130x faster first response time and 96.2 percent automated coverage across support tickets.
The solution runs on Amazon Bedrock AgentCore, integrating tools like Jira and Confluence alongside clinical regulatory security standards.
The team designed a skill library of 33 components and split tasks between deterministic Python code and large language models.
Strategic Analysis and Industry Implications
The clear separation between deterministic Python tasks and language reasoning enhances system transparency. This allows the engineering team to easily isolate technical faults when a diagnosis is incorrect.
Adopting a keyless deployment model through OpenID Connect and AWS STS minimizes credential exposure risks in demanding operational environments.
Outlook and Key Factors to Watch
It remains unclear how effectively SWARM generalizes to biological or medical equipment workflows outside Natera's initial testing scope.
Detailed operating costs and cloud resource requirements needed to sustain the dual-runtime architecture have not been fully disclosed.







