Principal AI Engineer — End-to-End SaaS & Systems

Northwoods builds software that empowers Health and Human Services professionals to better serve vulnerable families. We are hiring a Principal Engineer to carry dual accountability: deliver Traverse product features end-to-end, and co-build the agentic engineering framework that continuously raises the standard for how software gets built here.

This is a technically authoritative, principal-level role on a high-autonomy engineering team where your decisions set the standard and your output shapes the system. You will define the technical approach, execute against it, and course-correct independently. AI-forward development is not a tool you reach for occasionally; it is the operating model for everything you build and every process you improve.

Experience

Required:

• 8+ years of professional software development experience, with demonstrated delivery at principal or staff level in a SaaS product environment
• Full-stack proficiency across a modern web stack; direct experience with Ruby on Rails, React/Type Script, and Postgre
  
  SQL is highly valued
• Expert-level, hands-on AI-assisted development — not theoretical familiarity, but shipped production work built using agentic tooling (Claude Code, Cursor, or equivalent)
• Demonstrated ability to design systems from scratch, reason about architectural tradeoffs, and communicate decisions clearly to both technical and non-technical stakeholders
• Strong automated testing ownership: unit, integration, and end-to-end — you build tests as part of delivery, not after
• Must be authorized to work in the U.S.

What you’ll be doing

• Own and deliver Traverse product features end-to-end — from technical design through implementation, automated testing, deployment, and production operation — with full accountability for quality at every stage
• Set and enforce the code quality bar for the team: architecture decisions, design patterns, code review standards, and security-by-default practices that every engineer on the team follows
• Build and continuously improve the agentic engineering framework — automated test generation pipelines, AI-assisted code review workflows, prompt engineering standards, and agentic workflow runbooks that make the whole team faster
• Drive AI-forward engineering maturity — identify the next automation opportunity, write the runbook, ship it; this is ongoing delivery work, not a side project
• Mentor and technically develop the engineers on the team through pairing, code review, and direct, actionable feedback
• Operate with minimal supervision — define the technical approach, execute against it, and course-correct independently without waiting for direction

You may be a good fit for our team if you have the following skills…

• Expert-level AI-assisted development as a daily working practice — you run agentic workflows, manage context deliberately, and use Claude Code, Cursor, or equivalent to ship production work at accelerated pace
• Full-stack range without hard gaps — you are as comfortable in a Rails controller and Postgre
  
  SQL schema as you are in a React component and a CI pipeline
• Strong system design instincts — you can architect a feature or service from scratch, reason about tradeoffs, and explain decisions clearly to both engineers and non-technical stakeholders
• A high bar for code quality that you hold for yourself and others — clean code, SOLID principles, test-first thinking, and security-by-default are how you work, not aspirations
• Demonstrated ownership of automated testing end-to-end — unit, integration, and E2E coverage is part of your definition of done, built and owned by the engineer who wrote the feature
• A builder's mindset about engineering systems — you don't just follow the SDLC, you improve it; writing runbooks, automating processes, and raising the baseline is work you seek out
• High autonomy and low need for direction — given a problem, you define the approach, execute, and surface blockers early rather than waiting for a manager to unblock you
• Strong written communication — your design docs, code review comments, and architecture decisions are clear, concise, and useful to the person who wasn't in the room when you made them
• Comfort operating in a flat,…