Senior Software Engineer

ComPsych is a provider of organizational mental health, well-being, and absence management solutions. The Senior Software Engineer will help modernize the company's technology through AI-led development, focusing on system design, task decomposition, technical direction, testing, security, reliability, observability, and production operations. The role also coaches engineers on effectively using AI throughout the software development lifecycle.

Responsibilities:

• Plan and design as your core contribution. You own problem framing, sequencing, and tradeoffs with Product and Design, and the system design that follows — the data model, the API contracts, the failure modes
• Direct implementation rather than writing it line by line. Decompose the work, write the specs, and let AI agents build — dropping into the code yourself when debugging or complexity demands it
• Own testing as a design decision. You decide what needs coverage and at what level, direct AI to build it, and validate that the tests prove what they claim — not just that they pass
• Build security, reliability, and observability in from the start. They're design inputs, not production afterthoughts — the systems you ship can be operated safely and understood when they fail
• Review everything with the same rigor, regardless of author, and keep production judgment human. You're accountable for what ships whether you or an agent wrote it. Deployment calls, on-call response, and what an incident teaches you stay yours
• Raise the bar for how the team works with AI, coaching engineers to specify, delegate, and validate rather than hand-code everything themselves

Requirements:

• Deep server-side experience in a modern language such as Java, Python, C#, Kotlin, or Go
• Must have 5+ years of full stack development experience
• APIs and data models other people depend on: REST contracts, versioning, OAuth2/OIDC, schema design, and SQL that holds up under real load
• Async and event-driven systems in production — Kafka or comparable, idempotency, and a clear view of what at-least-once costs you downstream
• You've built and operated production systems at real scale on AWS or Azure — containers, Kubernetes or serverless, IaC, and CI/CD pipelines you've built, operated, or materially improved
• A testing strategy you can defend — what belongs at unit, integration, and contract level, and how you know a release is verifiable before it ships
• Reliability and security you've been accountable for: SLOs you've been paged against, incidents you led, OWASP and threat modeling applied — and PHI/HIPAA or comparable regulated-data experience
• Agentic AI is your primary development surface, and you catch what it gets subtly wrong. Not occasional autocomplete, not a hunch — you've shipped and operated real production systems working this way, and can point to what you caught before it shipped
• LLM application engineering (RAG, retrieval quality, evals, LLM observability), or modern TypeScript frontend work against a design system