Software Dev Engineer II, AWS Networking Reasoning Team

The AWS network changes constantly, and every change carries risk. A single bad configuration can pull capacity offline or disrupt customer traffic. Validating that a change is safe at AWS scale is hard: a single change interacts with thousands of variables across a global network, and the complexity grows faster than the size does. That challenge is now increasing in a new way. Network teams are building their own software services that manage network behavior directly, and they need to propose, test, and release a change in minutes opposed to weeks. Our team builds the platforms that let them do that without giving up reliability.

We build the software that tells a network builder whether a proposed change is safe before it reaches production. That means formal verification of device configurations, large-scale network simulation and emulation, and automated testing that models how the network will behave under a change. We turn those capabilities into self-service guardrails that any network team can adopt and trust, so they own their safety decisions and still move fast. The hard part is keeping the checks correct and general as the network evolves, and fast enough that builders rely on them instead of routing around them.

Amazon Web Services (AWS), Software Engineering, Network Administration/Management, Risk, Amazon Web Services (AWS), Test Automation, Network Simulation, Formal Verification, Software Development, Technical Leadership