We are hiring an Agentic AI Engineer to design and build intelligent AI systems that autonomously reason, decide, and execute actions across enterprise insurance workflows.
This role goes beyond model development. You will build agent-based systems that interact with carrier platforms (including Guidewire), interpret customer intent in real time, and trigger automated or human-in-the-loop actions across the insurance lifecycle.
This is a production engineering role focused on real-world deployment, system integration, and decision automation at scale.
Key Responsibilities
- Agentic AI System Development
- Design and build AI agent frameworks for real-time decisioning and workflow automation
- Develop multi-step reasoning systems using LLMs, tools, memory, and policy-driven logic
- Build orchestration layers that determine when to automate vs route to human agents
- Implement intent detection and action prediction pipelines using behavioral and interaction data
LLM / AI Engineering
- Expertise in Python and AI development using modern ML/LLM libraries
- Hands-on experience with OpenAI APIs and LLM integration frameworks
- Design and optimize prompts, tool-calling systems, and structured output pipelines
- Fine-tune and optimize model performance for latency, cost, and accuracy
- Experience with Agentic AI patterns (ReAct, tool use, multi-agent workflows)
Systems & Integration
- Integrate AI systems into enterprise platforms including Guidewire, Duck Creek, or Majesco
- Build APIs and middleware services connecting AI engines to core insurance systems
- Work with structured and unstructured insurance data (policy, claims, customer interaction data)
- Collaborate with Guidewire developers and enterprise integration teams
Collaboration & Delivery
- Work closely with product, engineering, and insurance domain teams to translate workflows into AI systems
- Participate in architecture design for scalable AI-driven insurance platforms
- Maintain technical documentation for agent workflows, APIs, and system behavior
- Ensure production-grade stability, monitoring, and observability of AI systems
Required Skills
- 4–8 years of software engineering or AI engineering experience
- Strong Python development expertise
- Hands-on experience with LLMs (OpenAI or equivalent)
- Strong understanding of AI agent frameworks and orchestration patterns
- Experience with API development and system integration
- Knowledge of NLP, embeddings, classification, or ranking systems
- Experience working with production systems, not just prototypes
- Strong Plus (Differentiators)
- Experience building agentic AI systems or multi-agent workflows
- Exposure to insurance platforms: Guidewire, Duck Creek, Majesco
- Experience with workflow engines or decision systems
- Experience with real-time data pipelines or streaming architectures
- Kubernetes / Docker / cloud-native deployment experience
- Understanding of customer lifecycle or CRM systems
Must-Have Mindset
- Strong bias toward shipping production systems, not research experiments
- Comfortable working in ambiguous, high-ownership environments
- Ability to translate business workflows into deterministic + probabilistic AI systems
- Strong debugging mindset for complex AI + system interactions
What Success Looks Like
- AI agents deployed into real carrier workflows
- Measurable impact on conversion, engagement, or operational efficiency
- Stable integration with enterprise insurance systems (including Guidewire ecosystems)
- Reduction in manual intervention through autonomous decisioning systems
- Reliable, observable, production-grade AI infrastructure
