Senior Software Engineer, Insurance

Porch Group is a vertical software and insurance platform focused on helping homebuyers move, maintain, and protect their homes. The Senior Software Engineer, Insurance will build and own backend services for underwriting, administering, and scaling insurance products, including AI-enabled platform capabilities. The role covers platform architecture, policy administration, migration tooling, production delivery, and collaboration with product and insurance specialists.

Responsibilities:

• Use agentic AI coding tools as your primary development methodology for the platform build, not as an experimental side channel
• Work within the team’s agent configuration patterns, custom skills, and CI/CD integration — and help improve them as you find rough edges
• Make sure the code you ship with AI assistance meets our standards: a green CI pipeline, passing evaluations, and clean pre-commit hooks are non-negotiable
• Apply good spec-driven, context-management practices that separate productive AI-assisted work from churn, and share what works with your teammates
• Build agentic capabilities into the platform — agent-assisted configuration, AI-driven underwriting decisioning, document understanding for declarations and endorsements, and conversational interfaces for the operations team
• Contribute to the RAG pipelines, retrieval, and tool-use patterns that ground LLM behavior in regulated insurance content (rate manuals, underwriting guidelines, state-specific rules)
• Implement the integration patterns (MCP, function calling, structured outputs) that expose platform capabilities to internal agents and external partners
• Add observability, evaluation, and cost instrumentation to the AI features you own so their behavior is measurable, debuggable, and bounded
• Follow the responsible-AI guardrails that keep agent-influenced decisions auditable and aligned with insurance regulatory requirements
• Design and build core PAS services in your area — quoting, rating, binding, issuance, endorsements, renewals, cancellations, and reinstatements — making well-reasoned design decisions and tradeoffs as you go
• Contribute to the rating engine, product configuration layer, and rules framework that let business teams launch new states and products without engineering bottlenecks
• Implement data models for policy lifecycle, exposure management, and regulatory compliance, with guidance from senior engineers on the broader architecture
• Build within the platform’s event-driven patterns and integration architecture for the carrier, vendor, and partner systems the platform speaks to
• Build migration tooling: automated data extraction, transformation, validation, and reconciliation pipelines
• Help keep existing vendor platforms stable while we build the replacement — you’ll context-switch between “keep the lights on” and “build the future” without dropping either
• Partner with Product and Insurance SMEs to refine requirements and turn domain needs into working software
• Participate in design reviews, and support teammates through constructive code reviews and ad-hoc technical advice
• Lead by example — strong ownership, accountability, and a high quality bar — and help other engineers grow their skills
• Stay hands-on. This is a build role: you write production code every week

Requirements:

• 5+ years of professional software engineering experience building and operating backend services in production
• Demonstrated ability to lead at least one project from conception to production, breaking work into reasonably sized, well-tested releases
• Production-grade engineering depth in Python, Java, or a comparable backend language; comfortable across the stack from data model to API to deployment
• Solid systems design fundamentals: distributed systems, event-driven architecture, API-first design, and data modeling for transactional workloads
• Hands-on experience building LLM-integrated features — RAG, tool use, or agent workflows — running in real software, not just prototypes
• Working familiarity with agentic AI coding tools as part of your day-to-day development, with a point of view on what makes AI-assisted work reliable
• Requires minimal direction to deliver in your area, and comfortable working in unfamiliar parts of the codebase with guidance
• US-based; able to work core business hours
• Exposure to homeowners or other personal-lines P&C insurance systems, or eagerness to ramp deeply on the domain
• Experience contributing to a custom-built policy administration system, or to a migration off a legacy/vendor PAS (data migration, parallel-run validation, cutover)
• Experience with AI evaluation, observability for agent runs, prompt and model versioning, or AI cost-management
• Working knowledge of Model Context Protocol (MCP), function calling, and structured outputs, and patterns for grounding LLMs in regulated content
• Familiarity with regulatory frameworks: SERFF rate filings, NAIC data calls, state DOI requirements, and the NAIC AI bulletin
• Working knowledge of insurance data standards (ACORD, MISMO) and the open-source PCDM data model
• Background in cloud-native infrastructure (GCP preferred), event streaming (Kafka), and modern observability practices
• Bachelor's degree in Computer Science, Engineering, or equivalent practical experience