AI Engineer - Customer Service

The Mission

We're building an AI-powered insurance brokerage that's transforming the $900 billion commercial insurance market by automating processes that currently run on pre-internet systems.

We just finished YC and are emerging from stealth with $8M in funding from an amazing bench of investors. We're looking for an exceptional AI Engineer who can build our end-to-end sales process and conversion systems from the ground up and drive measurable business outcomes.

You'll create systems that turn traditional insurance sales cycles into efficient, AI-driven processes that improve conversion at every step from lead to binded policy. This role sits at the critical intersection of engineering, AI, and sales—a position to directly impact revenue as we scale from seed to industry leader. You will work directly with the CEO and CTO to shape our sales automation strategy and technical roadmap.

Be bold & aggressive enough to believe we can scale from $1.2M to $100M in run-rate ARR by December.

Outcomes You'll Drive

• Build voice AI systems capable of sustaining natural, contextually rich conversations for 15+ minutes to handle complex customer service inquiries and insurance policy management
• Develop multi-channel AI agents that seamlessly manage customer interactions across voice, email, chat, and web interfaces
• Create sophisticated retrieval systems that instantly surface relevant policy information, coverage details, and customer history during interactions
• Implement automated workflows for common insurance transactions including policy changes, certificates of insurance, additional insureds, and claims initiation
• Design intelligent escalation systems that know when to involve human agents while providing them with complete context
• Build self-service customer portals that enable policy management, document retrieval, and transaction history tracking
• Develop analytics systems that measure customer satisfaction, resolution rates, and service efficiency
• Create continuous learning loops where every customer interaction improves future AI performance

You're Our Person If

• You've built customer-facing AI systems that successfully automated complex service workflows
• You have experience developing voice AI capable of extended (15+ minute) natural conversations with complex talk paths
• You're equally comfortable with TypeScript/Node.js and Python/ML frameworks
• You ship features daily and think in terms of weekly experiments, not quarterly plans
• You have a data-driven mindset and can trace your work directly to customer satisfaction and operational metrics
• You've implemented sophisticated RAG systems that retrieve and present complex domain knowledge
• You thrive at the intersection of engineering, AI, and customer experience
• You understand insurance operations or have experience in regulated industries
• You're proficient with modern coding environments like Cursor or WindSurf IDE for maximum productivity

Hard Requirements

• Strong experience with both TypeScript/Node.js and Python
• Experience with modern agent orchestration frameworks (Mastra, LangGraph) and Model Context Protocol (MCP)
• Experience building sophisticated voice AI systems capable of sustaining natural conversations for 15+ minutes with complex talk paths
• Proven track record implementing RAG systems for domain-specific knowledge retrieval
• Experience with multi-channel customer service automation (voice, email, chat, web)
• Experience integrating AI systems with backend transactional systems or CRMs
• Advanced usage of Cursor or WindSurf coding IDE
• Must be based in San Francisco and work in-office 5.5 days per week (relocation assistance provided)

Our Sales & Data Stack

We're building a modern, AI-native customer service infrastructure to power our growth:

Voice & Conversational Systems:

• LiveKit for real-time, bidirectional voice communication with ultra-low latency
• Twilio for enterprise-grade telephony infrastructure
• OpenPhone for capturing and analyzing human conversations
• Real-time intent recognition and conversation path optimization

Agent Orchestration:

• Mastra and LangGraph for sophisticated agent orchestration with parallel reasoning paths
• Model Context Protocol (MCP) for optimized LLM interactions and context management
• Custom-built service agents specialized for different policy types and customer needs
• Human-in-the-loop workflows for complex cases requiring expertise

Knowledge Retrieval & Management:

• LlamaParser for document parsing and information extraction from policies
• Sophisticated RAG architecture with hybrid lexical-semantic search capabilities
• Multi-vector retrieval for policy details, customer history, and insurance regulations

Memory & Storage:

• Zep AI for persistent graph-based conversation memory
• PostgreSQL for relational customer and policy data
• Redis for high-performance, low-latency operational memory
• Vector store with hybrid lexical-semantic search for insurance domain knowledge

Core Engineering:

• TypeScript/Node.js for robust application development
• Python for AI/ML workflows and advanced NLP
• React/Next.js for customer-facing self-service portals
• Temporal workers for distributed, fault-tolerant workflow orchestration
• Harper Hub, our operational platform with advanced workflow capabilities

What You'll Build in Your First 90 Days

First Month:

• Build and deploy the initial voice AI system capable of handling basic policy inquiries and service requests with 15+ minute conversation capability
• Implement the foundation for a multi-channel customer service platform (voice, email, chat)
• Create the initial RAG system for retrieving policy information and customer history
• Develop observability tools for monitoring conversation quality and customer satisfaction

Second Month:

• Expand voice AI capabilities to handle policy modifications, certificate requests, and claims initiation
• Build the customer self-service portal for policy management and document access
• Implement intelligent escalation paths that seamlessly transition from AI to human agents
• Create automated workflows for common insurance transactions and service requests

Third Month:

• Develop sophisticated analytics that measure service efficiency and customer satisfaction
• Build personalization systems that adapt interactions based on customer history and preferences
• Implement continuous learning mechanisms where successful resolutions improve future interactions
• Scale the system to handle increasing volume while maintaining quality and satisfaction
• Integrate with policy management and claims systems for end-to-end transaction processing

Customer Service Philosophy

• Conversation-First: Design natural voice and text interactions that create trust and satisfaction
• Knowledge-Driven: Ensure every interaction is backed by accurate, comprehensive policy information
• Resolution-Focused: Measure success by complete problem resolution, not just handling volume
• Human+AI Collaboration: Create systems where AI handles routine matters and humans add expertise
• Continuous Improvement: Build feedback loops that automatically enhance service quality over time
• Customer Experience: Design every interaction to enhance satisfaction and retention

We require you to be in San Francisco and work from our office 5.5 days per week. We'll cover relocation costs and believe the best growth teams collaborate intensively in person.