Applied AI Engineer, Enterprise

About Carrot: Carrot is the leading global fertility and family care platform, built on intelligent care orchestration: the right clinical guidance, at the right moment, in the context of each member’s life. More than a thousand multinational employers, health plans, and health systems trust Carrot to support millions of members across 195 countries – from pre-pregnancy through menopause and major life moments in between. Carrot's comprehensive clinical program delivers industry-leading cost savings for plan sponsors and award-winning experiences and improved outcomes for millions of people worldwide. Carrot is widely regarded as a defining force in healthcare innovation as a recipient of several top-tier awards, including Fast Company's 'Most Innovative Companies' and CNBC's '100 Barrier Breaking Startups'. The company is regularly cited by leading global outlets — including The Economist, Bloomberg, The Wall Street Journal, NPR, ABC News, and Harvard Business Review — as a leading voice on digital health, the future of work, and family health. Learn more at get-carrot.com . The Opportunity 🚀 Carrot Fertility is hiring an Applied AI Engineer to join our Enterprise Technology team. You will design, build, and ship production-grade AI and integration solutions that give internal teams reliable, structured access to Carrot's core product and operational data. This is a hands-on engineering role — you will own delivery end-to-end: from scoping and architecture through deployment, iteration, and measurable business impact. Your first project will be building the data access layer for Carrot's enterprise AI agent ecosystem — designing and deploying an MCP architecture that exposes structured, governed access to Carrot's core product and operational data. As Carrot's AI capabilities grow more sophisticated, they require deterministic and programmatic access to core operational data: member eligibility, benefit balances, expense records, provider information, employer-specific rules, and more. You will build that layer — cleanly, auditably, and in a way the broader team can maintain and extend. This is the kind of foundational, high-leverage infrastructure work you will take on regularly. You will be embedded with internal teams across Operations, Business Systems, and Product — translating data access needs and workflow gaps into AI-powered solutions that create lasting operational leverage. This is not a slow-start role. You will move with the urgency of a startup engineer and the judgment of a senior architect, while holding to a core design principle: least complexity. Build the right thing with the right tool, and build it in a way the team can maintain and extend long after you've moved on to the next problem. What You'll Do Embed directly with internal business teams to discover data access gaps and workflow pain points, prototype solutions rapidly, and own the full delivery lifecycle from scoping through production deployment. Architect and build agentic AI systems that handle complex, multi-step business processes — producing reliable, deterministic, auditable outcomes even in high-stakes or regulated contexts. Design systems with compliance and data governance baked in: HIPAA-compliant data handling, role-based access control, prompt hygiene, evaluation frameworks, and observability throughout. Write high-quality, production-grade code alongside platform-based integrations. You are comfortable choosing the right tool for the job — low-code where it reduces delivery time and maintenance burden, custom code where it provides control or capabilities that low-code cannot. This is not an exclusively low-code or exclusively greenfield engineering role — it is both, applied with judgment. Use Claude Code as your primary AI-assisted development environment, leveraging agentic coding deeply to accelerate delivery and maintain high output quality. Translate ambiguous business requirements into clean technical designs and communicate them clearly to both technical and non-technical stakeholders. Define and instrument success metrics for each solution: time saved, error rates, SLA improvements, cost reduction, and user satisfaction. Feed patterns, failure modes, and reusable frameworks back into a shared internal playbook to compound team velocity over time. Lead enablement: run knowledge-transfer sessions and create documentation so business teams can understand, trust, and extend what you build. Build and maintain MCP (Model Context Protocol) servers that expose Carrot's core product and operational data to AI tools and agents — spanning multiple data domains  Develop deep familiarity with Carrot's application database schema so you can independently navigate the data model, design reliable queries, and build clean access patterns without requiring constant support from Product Engineering. Build Claude skills and plugins that business teams and AI agents can use, extend, and maintain. Apply a design of least complexity principle to every build: prefer simple, maintainable, documented solutions over sophisticated ones, and build in a way that does not create single-person knowledge dependencies. Example Projects You Might Tackle 💼 Application data access layer and AI agent enablement (your first project): Design and deploy a suite of MCP servers that expose structured, governed access to Carrot's core product and operational data. Data domains include expense records, uploaded files, member eligibility and benefit dates, benefit balances, merchant and currency data, journey and phase, medical and clinical questions, care recipients, employer-specific rules, and pharmacy network access. Surface this MCP layer into Carrot's AI agent infrastructure and tune the agent to meaningfully improve the quality and reliability of AI-assisted internal operations. AI-assisted intake and routing: Build an intelligent triage system that classifies inbound request