Loading...
Software Dev Engineer, AWS AI Agentic Automated Reasoning (AR)
Amazon
On-siteseniorPosted May 17, 2026
About the Role
Design and implement features for an automated reasoning platform, build and maintain distributed systems, evolve service APIs, collaborate with scientists, and ensure service reliability.
Requirements
Seeking a Software Development Engineer to design, build, and operate cloud-hosted automated reasoning platforms, integrating solver improvements and ensuring operational excellence for AWS services.
Full Job Description
We are building exciting new capabilities in the Amazon Web Services (AWS) Agentic AI Automated Reasoning group by using Automated Reasoning in new, novel and exciting ways to enhance AWS services. The position is based in Seattle, Washington.
We are seeking a talented and passionate Software Development Engineer (SDE) who wishes to work at the intersection of Automated Reasoning and cloud-scale distributed systems, building and operating managed reasoning infrastructure that powers correctness guarantees across AWS.
As an SDE on the ATR team, you will design, build, and operate components of Amazon's cloud-hosted automated reasoning platform that enables AWS services to offload satisfiability and verification queries at scale. You will tackle challenges in distributed systems, solver integration, API development, and service reliability. Your work will directly enable customers to verify correctness of software, infrastructure configurations, and AI-generated code at AWS scale.
This is a unique opportunity to combine strong software engineering skills with automated reasoning technology. You'll work alongside world-class scientists and senior engineers to turn research breakthroughs in solver technology into production services that power correctness guarantees for millions of customers. The problems are hard, the impact is enormous, and your contributions will be highly visible.
Key job responsibilities
- Design and implement features and components of Amazon's cloud-hosted automated reasoning platform, including query routing, solver orchestration, result caching, and soundness management.
- Build and maintain distributed systems that meet AWS's operational excellence standards for availability, latency, and correctness.
- Contribute to the evolution of the service API, making it easier for customers to integrate reasoning capabilities into their workflows.
- Collaborate with Automated Reasoning scientists to integrate solver improvements and new capabilities into the production platform.
- Participate in on-call rotations, build monitoring and alerting, and contribute to the continuous improvement of service reliability.
- Write clean, well-tested, production-quality code and participate actively in code reviews and design discussions.
- Work with internal AWS customer teams to understand their use cases and help troubleshoot integration issues.
A day in the life
A typical day may include implementing a new feature for the solver service, collaborating with a scientist on integrating a solver optimization, participating in code reviews, debugging an issue in the query pipeline, or writing tests to improve coverage. You'll spend most of your time writing code and solving technical problems, while also contributing to design discussions and operational improvements. You might also attend a sprint planning session, pair with a teammate on a tricky problem, or investigate a customer-reported issue.
We are seeking a talented and passionate Software Development Engineer (SDE) who wishes to work at the intersection of Automated Reasoning and cloud-scale distributed systems, building and operating managed reasoning infrastructure that powers correctness guarantees across AWS.
As an SDE on the ATR team, you will design, build, and operate components of Amazon's cloud-hosted automated reasoning platform that enables AWS services to offload satisfiability and verification queries at scale. You will tackle challenges in distributed systems, solver integration, API development, and service reliability. Your work will directly enable customers to verify correctness of software, infrastructure configurations, and AI-generated code at AWS scale.
This is a unique opportunity to combine strong software engineering skills with automated reasoning technology. You'll work alongside world-class scientists and senior engineers to turn research breakthroughs in solver technology into production services that power correctness guarantees for millions of customers. The problems are hard, the impact is enormous, and your contributions will be highly visible.
Key job responsibilities
- Design and implement features and components of Amazon's cloud-hosted automated reasoning platform, including query routing, solver orchestration, result caching, and soundness management.
- Build and maintain distributed systems that meet AWS's operational excellence standards for availability, latency, and correctness.
- Contribute to the evolution of the service API, making it easier for customers to integrate reasoning capabilities into their workflows.
- Collaborate with Automated Reasoning scientists to integrate solver improvements and new capabilities into the production platform.
- Participate in on-call rotations, build monitoring and alerting, and contribute to the continuous improvement of service reliability.
- Write clean, well-tested, production-quality code and participate actively in code reviews and design discussions.
- Work with internal AWS customer teams to understand their use cases and help troubleshoot integration issues.
A day in the life
A typical day may include implementing a new feature for the solver service, collaborating with a scientist on integrating a solver optimization, participating in code reviews, debugging an issue in the query pipeline, or writing tests to improve coverage. You'll spend most of your time writing code and solving technical problems, while also contributing to design discussions and operational improvements. You might also attend a sprint planning session, pair with a teammate on a tricky problem, or investigate a customer-reported issue.