Applied Scientist, Automated Reasoning Group

Amazon Inc.; Seattle (WA); On-site; USD 150,000 - 210,000; Full time

Job summary

Amazon Inc. in Seattle is seeking an Applied Scientist for the AWS Automated Reasoning group to develop and apply automated-reasoning techniques, formal methods, and neurosymbolic approaches to ensure safety, security, and correctness of code and services.

You will collaborate with leading scientists, publish findings, mentor teammates, and help shape technology, product, and customer impact across cloud infrastructure verification, software security, and policy validation.

Responsibilities

• Design and implement algorithms and formal methods for automated reasoning to verify correctness, security, and reliability of computing systems.
• Tackle complex problems, own design, implementation, and delivery with long-term impact.
• Develop new decision procedures, heuristics, and search strategies for scalable verification tools.
• Explore generative AI and ML techniques to enhance automated reasoning.
• Develop techniques for safe, aligned autonomous agents and verified AI outputs.
• Conduct original research and publish findings in peer-reviewed venues.
• Collaborate with customer teams to address verification needs with novel analysis capabilities.
• Provide cross-organizational technical influence and translate capabilities into production systems.
• Mentor scientists and engineers on formal methods and neurosymbolic techniques.

Job description

This is an Applied Scientist position in the AWS Automated Reasoning group. In this role you will work together with leading scientists in the field to develop and apply the latest automated-reasoning techniques, formal methods, and neurosymbolic approaches to ensure safety, security, and correctness of code and services. As Agentic AI becomes the primary method of software development, we are changing the way that computer systems are developed and operated, raising the bar on quality, verifiability, and durability.

Automated Reasoning at Amazon is central to maintaining customer trust and improving the customer experience. Automated Reasoning techniques are being applied in diverse areas spanning generative-AI safety, agentic sandboxes, cloud infrastructure verification, software security and verification, cryptographic assurance, and policy validation. The techniques we are applying include automated theorem proving, model checking, constraint solving, and neurosymbolic proof search.

This is a unique opportunity to get in early on a rapidly expanding part of AWS and help shape the technology, product, and business. You will have a chance to utilize your deep technical expertise within a fast-paced environment and make a large business and customer impact.

Key job responsibilities

• Design and implement algorithms and formal methods for automated reasoning, including constraint solving, model checking, static analysis, theorem proving, and program synthesis to verify the correctness, security, and reliability of computing systems.
• Solve large or significantly complex problems that require deep knowledge and scientific innovation in your domain; own strategic problem solving and take the lead on design, implementation, and delivery of solutions with long-term quantifiable impact.
• Develop new decision procedures, heuristics, and search strategies that improve the scalability and accuracy of verification tools; build and deploy production-grade automated reasoning systems at Amazon scale.
• Explore and apply generative AI and machine learning techniques to enhance automated reasoning capabilities, including learning-based heuristics for search and optimization, neural approaches to symbolic reasoning, and methods for verifying the correctness of AI-generated code.
• Develop automated reasoning techniques for generative AI and agentic coding systems, including methods for ensuring the safety and alignment of autonomous software agents and applying formal guarantees to large language model outputs.
• Conduct original research and publish findings in peer-reviewed venues.
• Work with customer teams to understand the nature of their software and the properties they need to establish; identify tools and methods capable of addressing verification needs, including novel analysis capabilities.
• Provide cross-organizational technical influence, increasing productivity and effectiveness by sharing deep knowledge and experience; collaborate with partner teams to translate verification capabilities into production systems.
• Mentor scientists and engineers on formal methods, neurosymbolic techniques, and best practices for building reliable automated reasoning systems; assist in career development of others.