Software Engineer 4 - AI Native

This position is listed on behalf of a partner company, who manages all applications and next steps. Our partner is looking for a Software Engineer 4 - AI Native based in United States.

This is a full-time, fully remote engineering opportunity within an AI-native software development lifecycle. The role focuses on directing autonomous coding agents, rigorously reviewing their output, and owning the quality of production software. You will work across high-volume development workflows using generator-verifier architecture, evaluation-driven development, staged deployment, and autonomous pull-request pipelines. The position combines strong software engineering fundamentals with hands-on experience orchestrating AI agents in production environments. You will help define measurable quality standards while maintaining responsibility for correctness, performance, security, and accessibility. Working within a FedRAMP-authorized environment, you will also contribute to evolving how engineering teams safely and effectively operate with AI at scale.

Accountabilities:

• Direct autonomous coding agents across production workstreams by decomposing complex engineering problems into suitable tasks, dispatching agent workflows, reviewing their output, and integrating approved work into production software.
• Review agent-generated pull requests at both high volume and technical depth, identifying correctness, security, accessibility, and other defects that automated testing may not detect while maintaining a consistent engineering quality bar.
• Design, author, and maintain evaluation suites that establish measurable standards for software quality, using evaluation-driven development to validate correctness rather than relying solely on subjective assessment.
• Own end-to-end quality for software delivered by the team, including correctness, performance, security posture, reliability, and WCAG accessibility regardless of whether the initial implementation was produced directly by an engineer or an autonomous agent.
• Advance engineering work along an autonomy ladder based on measurable reliability, increasing autonomous execution when evidence supports it and reverting to more supervised workflows when reliability does not meet established standards.
• Identify weaknesses in prompts, development patterns, evaluation processes, and pipeline components that emerge at scale, collaborating with technical leadership to improve the AI-native development lifecycle.
• Maintain autonomous agent operations within established security boundaries, including branch-only execution, vaulted credentials, sandboxed actions, and in-VPC inference, without compromising security or compliance for increased throughput.
• Contribute to engineering standards and reusable prompt and evaluation patterns, while mentoring engineers transitioning to agent-directed software delivery and helping expand proven practices across development teams.
• Support a secure and compliant engineering environment by treating requirements such as NIST 800-53 Rev. 5, WCAG, SOC 2, and applicable FedRAMP authorizations as integral engineering requirements.

Requirements:

• Strong software engineering fundamentals, including data structures, systems design, testing, and the ability to quickly understand unfamiliar code and accurately evaluate implementation quality.
• Demonstrated experience shipping production software while taking ownership of its quality, including diagnosing failures, identifying root causes, and implementing effective remediation.
• Hands-on experience directing coding agents in production environments, including a practical understanding of agent failure modes and the ability to critically evaluate agent-generated code rather than approving it by default.
• Proven code-review expertise, with the ability to identify defects that automated tests may miss, provide actionable feedback, and maintain a high quality standard without becoming a bottleneck.
• Strong autonomy and ownership, with the ability to advance complex work without step-by-step direction, make sound engineering decisions, and proactively escalate risks or issues.
• Experience in high-assurance or regulated environments such as FedRAMP, defense, financial services, healthcare, or other NIST 800-53, SOC 2, or HIPAA-bound environments is preferred.
• Experience authoring evaluation suites or working in rigorous test-first or evaluation-driven development environments is preferre…