Sunlight

About Sunlight Sunlight is a mission-driven technology company building innovative solutions that connect child-welfare-involved parents and youth with their legal counsel, caseworkers, mentors and other support resources. Our secure, real-time communication platform promotes trust and transparency across the child welfare system and directly contributes to shortening the time children spend in foster care and helping families reunify safely and sooner. About the role At Sunlight, we've cracked velocity. Our agentic development pipeline ships features fast — sometimes hundreds or thousands of lines per feature cycle. We've built the engineering backbone that makes that velocity trustworthy, scalable, and secure — and now we need a senior engineer to take ownership of making it even better. This isn't a traditional engineering role — it's much more. You'll be shaping the future of AI-assisted software development — pushing the boundaries of what agentic engineering can do, optimizing how it works, and building the processes that let a small team deliver like a big one. You need the skillset of a team lead — because you are one. Your team just happens to be AI agents, not humans. You'll orchestrate how they work, improve how they perform, and hold them accountable for the quality of what they produce. Role: Senior AI Quality EngineerType: Full-time · RemoteStack: Ruby on Rails · PostgreSQL · iOS · Android · Full-stackAI Tooling: Claude Code · Wispr FlowContext: AI-native engineering workflow · HIPAA-compliant legal techThe role in plain terms Think of this as being the lead of an AI engineering team — part architect, part process designer, part team lead whose reports happen to be AI agents. You'll orchestrate how work gets planned, executed, and validated. You'll review implementation plans before any agent writes a line of code. You'll validate output. And critically, when you see the same mistake twice, you'll fix the process so it doesn't happen a third time. This is a leadership role from day one. Not people management — agent management. Can you figure out how to orchestrate a team of AI agents to build a feature? Can you adopt and perfect processes that allow us to deliver quality work at scale, so we can focus all of our attention on what the software should do — not how it needs to be built? That's the job. Your legacy here won't be features you shipped — it'll be an engineering system that compounds in quality and efficiency because of the way you shaped it. What you'll actually do day to day - Review AI-generated implementation plans before code is written — catch architectural decisions that will cause problems downstream - Validate AI output across the backend: integrations, database performance, security, scalability, and infrastructure efficiency - Automate testing infrastructure across the full stack — build the systems that run unit, integration, acceptance, end-to-end, and production monitoring without manual intervention - Build feedback loops so the AI knows when something passes, when something breaks, and when to stop and bring a human in - Identify recurring failure patterns and improve the AI's prompting, agent configuration, or testing scaffolding to prevent them - Drive code quality standards — separation of concerns, clean architecture, testability — that get built into the AI's defaults - Own backend reliability: integrations run correctly, data flows are secure, performance scales, nothing leaks - Optimize the agentic development process itself — just because something can be done with a billion tokens doesn't mean it should be. Find ways to ship the same quality with less waste - Help shape and continuously improve the engineering spec process: take product specs, ensure the implementation plan is sound, and build repeatable processes that the team can scale on - Use our existing agentic orchestration to manage agent workflows end to end — plan the work, review the output, and continuously improve the system - Build and refine the processes that let our agentic pipeline operate with increasing autonomy and decreasing error rates - Own technical documentation that informs future AI sessions — so every agent starts with an accurate, up-to-date understanding of what it should be doing What you'll own - Technical Issues — triaging, validating, and resolving issues surfaced by AI agents or production monitoring - Performance Issues — identifying bottlenecks, optimizing queries, ensuring the system scales - Integrations — making sure third-party services and systems run correctly and reliably - Scalability — ensuring architecture decisions support growth without rework - Code Quality — setting and enforcing standards that get built into the AI's defaults - Architecture Quality — reviewing implementation plans for soundness before agents execute - Security — validating compliance within a HIPAA-compliant environment - Technical Documentation — owning the documentation that AI agents consume to understand the system. AI-generated and AI-consumed — this documentation directly informs the quality of future sessions - CI/CD Pipeline — health, optimization, and automated test run infrastructure What we're looking for Must have - Strong Rails background with real production experience - Test-driven development mindset — tests first, not tests last. This is critical to agentic development: if the AI doesn't have tests to validate against, it can't run autonomously - Experience designing testing architecture, not just writing tests - Ability to do high-volume code review: spotting patterns and validating whether AI-flagged issues are real problems or false positives - Comfortable working across the stack: web, mobile, APIs, infrastructure - Strong understanding of security best practices, especially in regulated environments - Self-directed — you identify what needs fixing and go fix it without waiting to be told - Excited about pushing the boundaries of AI-assisted development Nice to have - Experience with CI/CD pipeline design and optimization - Familiarity with Sentry, Pendo, or similar observability tools - Background in tech lead, team lead, or architectural review roles — you need this skillset now, not eventually - Experience managing engineering workflows, whether human or automated - Experience working alongside code generation or automation tooling (Claude Code, Cursor, or similar) - Query optimization and database performance experience - HIPAA, SOC 2, or similar compliance experienc The mindset that will make you great here - You care more about the system than the feature - You see a test failure as information, not inconvenience - You think in patterns, not just individual bugs - You make your own job easier over time by improving the process - You're excited to work with AI and want to push the boundaries of agentic development - You can give clear, structured feedback that improves the pipeline - You believe great software for a great cause is worth building with everything you've got - You think in orchestration, systems, and leverage; not lines of code written Why this role is different This is a new kind of engineering role. You won't be measured by features shipped or lines written. You'll be measured by the reliability, security, and efficiency of everything our AI produces — and how much better the entire system gets over time because of you. The mission matters. We need people who want to build great software for a great cause — and who see AI-assisted development as the way to punch above our weight as a small team doing important work. If you've been waiting for a place where engineering discipline actually compounds into something bigger than one codebase, this is it. Compensation Compensation at Sunlight Group is determined based on a combination of factors, including the scope and responsibilities of the role, the candidate's experience and skills, and internal equity across our team. The salary range listed for this position reflects our intended hiring range; offers are made thoughtfully within that range based on the full picture of a candidate's background. We believe in paying people fairly and are happy to discuss compensation openly throughout the hiring process.

• Use and continuously improve our existing agentic orchestration to build and iterate on features end to end. • Plan the work, direct the agents, review the output, and improve the process • Learn our codebase and domain, and help tighten and refine the platform as it exists today • Take new product ideas and feature requests from concept to implementation plan to deployable code — creating specs through voice-first, AI-assisted workflows, not traditional writing • Implement and manage product observability: track user behavior, click paths, time-on-task, drop-off points • Use AI tools to dogfood the app — surface areas where things look difficult, take too long, or require too many steps, and suggest improvements • Improve AI's ability to test the platform for usability and troubleshoot technical issues • Work with our personas and user research to ensure features serve the people who actually use the product — attorneys, CASA volunteers, caregivers, families

Sunlight is building systems that strengthen families and improve outcomes in child welfare and public sector work. That kind of mission doesn’t fit neatly into job descriptions — and neither do the best people. If you believe in what we’re building but don’t see a perfect match among our open roles, we’d still like to hear from you. Sometimes the right hire creates the role. Who Thrives Here You might be a strong fit if you: - Care deeply about improving systems that affect real families - Think in terms of outcomes, not just tasks - Operate with high ownership and low ego - Are comfortable building structure where it doesn’t yet exist - Can move between strategy and execution without drama We’re particularly interested in people with experience in: - Public sector SaaS - Child welfare, family defense, or court systems - Implementation & client operations - Data & analytics in human services - Partnerships & government relations - Product strategy in mission-driven tech - Nonprofit or public policy leadership But if your background doesn’t fit neatly into a box and you believe you can create disproportionate impact here, tell us why. What to Send Submit your resume along with a short note that answers: - What problem in our space do you care most about? - What are you unusually good at? - If we created a role for you, what would it be? If there’s alignment, we’ll start a conversation. We move thoughtfully, but we don’t miss exceptional people.