Software Engineer 4, AI-Native

The Company Serving the People Who Serve the People Granicus is driven by the excitement of building, implementing, and maintaining technology that is transforming the Govtech industry by bringing governments and its constituents together. We are on a mission to support our customers with meeting the needs of their communities and implementing our technology in ways that are equitable and inclusive. Granicus has consistently appeared on the GovTech 100 list over the past 5 years and has been recognized as the best companies to work on BuiltIn. Over the last 25 years, we have served 5,500 federal, state, and local government agencies and more than 300 million citizen subscribers power an unmatched Subscriber Network that use our digital solutions to make the world a better place. With comprehensive cloud-based solutions for communications, government website design, meeting and agenda management software, records management, and digital services, Granicus empowers stronger relationships between government and residents across the U.S., U.K., Australia, New Zealand, and Canada. By simplifying interactions with residents, while disseminating critical information, Granicus brings governments closer to the people they serve—driving meaningful change for communities around the globe. Want to know more? See more of what we do here. Job Summary Granicus  serves more than 7,000 public-sector agencies and powers approximately 30 billion digital interactions annually. Within the Office of the CTO, we operate an AI-native software development lifecycle: a production engineering model in which autonomous agents perform high-confidence implementation work and senior engineers orchestrate, review, and own the outcome. The model is established and operating; we are scaling it across additional delivery teams.  This role joins one of those teams. Our engineering standards are calibrated to the practices used by leading software and AI organizations —  generator-verifier architecture ,  eval-driven development ,  staged deployment , and  high-volume autonomous pull-request pipelines  — operated within a FedRAMP-authorized environment. We are hiring practitioners who will set and uphold these standards, not engineers who simply use AI tooling.  Operating constraints (non-negotiable):  agents execute only within branches; all agent-generated code passes senior human review before merging to production; autonomous execution is a force multiplier and does not transfer accountability away from the responsible engineer. Compliance obligations — NIST 800-53 Rev 5, WCAG, SOC 2, and applicable FedRAMP authorizations — are treated as engineering requirements and a source of competitive advantage.  Why this role exists   When agents produce the majority of initial implementation, the scarce engineering skill shifts from authoring code to directing it, reviewing it rigorously at volume, and owning whether it is correct. This role is for a senior engineer who has made that transition: decomposing work for the agent array, authoring the evaluation suites that hold agent output to a measurable standard, reviewing agent-generated pull requests with greater rigor than most engineers apply to their own work, and shipping production code within a FedRAMP-authorized environment. The role contributes to defining how engineers operate in an AI-native lifecycle, not merely to adopting AI tooling.  What Your Impact Will Look Like Direct the agent array on production workstreams  — decompose problems into tasks suitable for agent execution, dispatch them, and integrate the output into shipped software.  Review agent-generated pull requests at volume and at depth  — identify correctness, security, and accessibility defects that automated tests do not catch, while maintaining review throughput and a consistent quality bar.  Author evaluation suites that make quality measurable  — define criteria under which the pipeline validates correctness rather than relying on subjective assessment. Eval-driven development is your standard practice.  Own quality end to end  — correctness, performance, security posture, and WCAG accessibility of the software your team ships, irrespective of which component or agent produced the initial implementation.  Advance workstreams along the autonomy ladder on the basis of evidence  — move work from supervised to autonomous execution when measured reliability supports it, and revert promptly when it does not.  Strengthen the development lifecycle itself  — identify where patterns, prompts, or pipeline components degrade at volume and work with the lead architect to remediate them.  Maintain agent operations within the security boundary  — branch-only execution, vaulted credentials, sandboxed actions, and in-VPC inference. Throughput does not justify exceptions.  OWNERSHIP FROM DAY ONE An active workstream directed through the agent array Agent pull-request review as a core, high-signal responsibility Evaluation suites authored and owned for your team’s deliverables End-to-end quality ownership within the FedRAMP-authorized environment SCOPE YOU WILL GROW INTO Higher autonomy ratios substantiated by reliability data Reusable prompt and evaluation patterns adopted by other engineers  Contribution to the lifecycle standards as a co-author, not only a user  Mentorship of engineers transitioning to agent-directed delivery  You Will Love This Job If You Have Required   Strong engineering fundamentals.  Data structures, systems design, and testing, with the ability to read unfamiliar code quickly and assess it accurately. Agents amplify engineering judgment; they do not substitute for it.  A record of shipping production software you owned the quality of,  including responsibility for diagnosis and remediation when it failed.  Hands-on experience directing coding agents on production work,  including their failure modes and the practice of revie