Our client is a US-based, technology-enabled insurance brokerage platform that empowers entrepreneurs to establish and grow their own independent insurance agencies. The company provides access to insurance carriers, technology, marketing, and centralized operational support, allowing agency owners to focus on sales and customer relationships.
The company is currently investing in agentic AI solutions to automate manual insurance workflows, including policy changes, renewals, and other operational processes.
The existing platform includes a basic RAG solution, and the next step is to build more advanced, production-grade agentic AI systems.
Design, develop, and implement production-grade agentic AI systems for customer-facing and operational workflows.
Build and integrate agent orchestration, tool calling, memory, and context management capabilities.
Work on LLM model selection and optimization to improve system performance, reliability, and efficiency.
Evaluate and integrate existing AI frameworks, models, and tools into a cohesive agentic architecture.
Develop AI agents capable of executing complex, multi-step business workflows.
Collaborate with technical and business teams to translate insurance processes into agentic AI solutions.
Contribute to architectural decisions and provide hands-on technical leadership.
Strong software engineering background with experience designing and delivering complex production systems.
Hands-on experience designing and building production-grade agentic AI systems in insurance, fintech, or closely related financial services domains.
Experience implementing agents that autonomously or semi-autonomously execute business workflows, beyond simple chatbots or RAG-based question-answering systems.
Strong understanding of agent orchestration, tool calling, memory, context management, and stateful workflows.
Experience building or integrating agent frameworks, execution harnesses, and evaluation infrastructure.
Understanding of production AI system requirements, including observability, security, error handling, and human oversight.
Ability to independently evaluate technologies, make architectural trade-offs, and lead implementation.
