Architect, build, and ship full-stack features across our React/TypeScript frontend and Supabase/PostgreSQL backend.
Write production-quality code while moving at startup speed, using AI coding tools to dramatically compress the time between idea and shipped feature.
Make and defend architectural decisions: data modeling, API design, security, performance, scalability.
Maintain code quality, testing standards, and documentation without being told to.
Design and implement AI agents and multi-step automation pipelines that power our compliance engine: document ingestion, payroll validation, issue detection, remediation workflows.
Work with LLMs (currently Google Gemini and Claude) to build intelligent extraction, classification, and validation systems.
Build robust, observable, production-grade AI workflows with proper error handling, retry logic, audit trails, and human-in-the-loop checkpoints.
Establish patterns and practices that will scale as we grow the team.
Own security rigorously: data handling (PII, SSNs, payroll data), authentication, authorization, and audit logging are core to our value proposition.
Instrument the system: logging, monitoring, error tracking, performance.
Eventually recruit, mentor, and lead additional engineers as the team grows.
Requirements
Genuinely AI-native engineer: you live in Cursor, Windsurf, Claude Code, or equivalent. AI tooling is how you build, not a novelty.
Strong full-stack capability: React/TypeScript on the frontend, Node/TypeScript or equivalent on the backend, comfortable in SQL and relational data modeling.
Experience building AI agents and automation workflows in production -
not toy demos, but systems with real users and real consequences.
Architectural judgment: you know when to move fast and when to slow down, when to abstract and when to hardcode, when to build and when to buy.
Security-minded by default. Our customers share payroll data, SSNs, and financial records.
Independent operator who needs direction on priorities, not on how to execute.
Experience with Supabase, PostgreSQL, and serverless/edge function architectures (Deno a plus).
Demonstrated history of shipping meaningful software at startup pace.
Experience with document processing pipelines: parsing PDFs, Excel, Word documents; working with structured and unstructured data.
Comfort working in a domain you do not initially know deeply. Compliance and regulatory logic is learnable; the engineering fundamentals are not.