[Remote] Senior Full Stack Engineer

Note: The job is a remote job and is open to candidates in USA. Stratus is a company that provides advanced MEP specific solutions to optimize contractor workflows. They are seeking a Senior Full Stack Engineer to deliver customer value through feature work and operational support in their SaaS platform, utilizing AI-assisted development tooling as a primary part of the development process.

Responsibilities

• Ship features and customer-facing enhancements across the full stack, working from product specs and customer feedback
• Use AI-assisted development tooling (Claude Code, Cursor, Codex, etc.) as a first-class part of the dev loop — not as autocomplete, but as the primary pattern by which you produce, review, and validate code
• Build, share, and refine internal workflows that compound your velocity and the pod's — custom MCP servers, code-gen agents, automated review or test pipelines, codebase-exploration agents
• Validate AI-generated code before it ships — write tests, run it, review it, and exercise sharp judgment about when output ships, needs rework, or should be thrown away
• Investigate and fix customer-reported bugs in your area of the product, including in code you didn't write — including code an agent wrote
• Ramp quickly on unfamiliar codebases using AI as a working tool — design documents, code search, behavior reproduction
• Contribute to monolith decomposition and modernization as touch points emerge in your work
• Implement and maintain modern automated testing — unit, integration, and the scaffolding that AI-generated changes warrant
• Collaborate with cross-functional teams to define, design, and ship; coordinate deployments and incident response with peers
• Advocate for incremental delivery and contribute to the engineering culture that surrounds you

Skills

• 6+ years of professional software development experience, including production ownership of customer-facing features
• Polyglot full-stack engineer. You've shipped production code across multiple languages and stacks, and you ramp on a new one in days, not months. Language is a tool, not an identity. Comfort with at least one statically-typed backend language, at least one modern frontend framework, and both NoSQL and SQL data stores is assumed
• AI-assisted dev tooling is your dominant dev loop. You use Claude Code, Cursor, Codex, Copilot, or an equivalent daily — and you can describe specifically how your work changes because of it. 'I use Copilot for autocomplete' is not what we're hiring
• You have built workflows that compound your own velocity. Custom MCP servers for your dev environment, code-gen agents tailored to your stack, automated review/test/refactor pipelines, codebase-exploration agents — concrete examples expected. We are looking for engineers who have already moved past using AI on the side
• Sharp judgment about AI output. You can describe specific instances where you identified hallucinated or wrong AI-generated code and rejected it. You have a personal validation discipline — tests, runs, review — for everything an agent produces before it ships
• Hands-on experience with at least one mainstream agent framework or SDK (Claude Agent SDK, OpenAI Agents SDK, LangGraph, Pydantic AI, Mastra, or equivalent), even if only for personal or internal dev tooling. We want an opinion about strengths and weaknesses, not just a name-drop
• Prompt engineering at a senior level. You debug a misbehaving agent — including the agents in your own dev loop — by reading the trace, not by guessing. You've used structured outputs, tool-call schemas, and reasoning models in production, whether for your own workflows or for features you've shipped
• Eval-first mindset applied to AI-generated code. AI output doesn't ship until you've validated it. You have a personal discipline for this — automated tests, runs, ship/no-ship judgment, regression checks when you change your workflows. Familiarity with eval tooling (Braintrust, LangSmith, Langfuse, Inspect, OpenAI Evals, or equivalent) is welcome; the discipline matters more than the tooling
• Strong debugging skills — you can trace issues through complex systems, reproduce edge cases, and identify root causes in code you didn't write, including code an agent wrote
• Fast ramp on unfamiliar codebases with AI as a working tool — you understand how a system works by exploring it, not by waiting for documentation
• Working understanding of CI/CD pipelines and optimizing DevEx for a dev team
• Understands SOLID, DRY, KISS, YAGNI, and TDD and applies them when warranted
• Excellent communication; can work directly with product managers, designers, and customers to triage and scope work
• Working understanding of security best practices
• Working production knowledge of Model Context Protocol (MCP) — has built or extended MCP servers (for your own dev environment, your team's, or a customer surface)
• Multi-agent orchestration in personal dev tooling — e.g., orchestrating sub-agents for different review, test, or refactor passes
• Experience deconstructing monoliths into modular or service-oriented architectures
• Understanding of event-driven architectures and message-based integration
• Production experience with reasoning models, structured outputs, and tool-call schemas — for your own dev workflows or for shipped features
• Has shipped customer-facing AI features — welcome context, but not the primary signal we're hiring on for this role

Benefits

• Comprehensive and competitive health benefits plan
• Matching 401k contributions
• 20 days annual PTO
• Primarily remote work with occasional annual team onsites.

Company Overview