AI Engineer- Applied AI

Guidewire Software

About The Position

Guidewire is hiring an AI Engineer to build, evaluate, and harden Concierge, its production omnichannel AI agent for property and casualty (P&C) insurers. Concierge handles policyholder calls after accidents, gathers first notice of loss (FNOL) information, asks relevant questions, and records claims in ClaimCenter. The role focuses on agent orchestration, context architecture, tools, and evaluations that establish whether changes improve the system. Because the product operates over synchronous voice and handles regulated claim intake, latency and accuracy are critical. The team expands agent autonomy incrementally, relying on evidence to guide each increase. The engineer will own end-to-end slices of the agent, including prompt and context architecture, retrieval over policy and coverage data, calls to claims systems, and handoff to human agents. They will develop evaluation processes, analyze real production call traces, and use evaluation results to gate releases. They will improve containment and resolution while maintaining accuracy, instrument the system to measure results, and work on voice turn-taking, interruption, latency, and telephony infrastructure. The role also involves listening to call recordings, participating in carrier calls, and contributing product and insurance-context judgment to decisions about what the agent should do next. Guidewire describes this as a full-time position. The US base salary range is not provided in the posting. Eligible full-time positions and part-time roles of at least 30 hours per week receive benefits including health, dental, and vision insurance, paid time off, and a company-sponsored retirement plan. Some roles may also qualify for an annual bonus, commissions, and/or long-term incentive awards, subject to applicable factors.

Requirements

  • Experience shipping an LLM application that real users depend on.
  • Fluency in LLM application building blocks, including context engineering, retrieval-augmented generation (RAG), and agentic workflows, supported by real machine learning and deep learning knowledge.
  • Ability to make unpredictable systems sufficiently reliable through disciplined evaluations, error analysis, and statistical measurement.
  • Software engineering fundamentals, including reasoning about cost, latency, reliability, and scalability tradeoffs.
  • Experience designing data stores and distributed systems, writing tests, and taking security and privacy seriously.
  • Advocacy for CI/CD process automation.
  • Ability to use coding agents effectively, including managing context, deciding when to write specifications, providing verifiers, and choosing when to intervene.
  • Product sense and enough insurance context to help shape requirements and make decisions about the appropriate pace of delivery.
  • Commitment to continual learning as AI tooling and practices evolve.

Nice To Haves

  • Experience with real-time or streaming voice systems, including ASR, TTS, telephony, or SBC/SIPREC.
  • Experience in insurance, claims, or another regulated domain where errors have a cost.
  • Experience building internal evaluation or observability tools adopted by other engineers.
  • Experience with machine learning principles, model experimentation, and training and deploying models in production.
  • Experience in P&C insurance or exposure to Guidewire products and services.
  • Experience applying prompt or pipeline optimization frameworks such as DSPy or GEPA to a production system.

Responsibilities

  • Build and maintain end-to-end parts of the agent, including prompt and context architecture, retrieval over policy and coverage data, tool and API calls into core claims systems, and handoff to a human when appropriate.
  • Build and run the evaluation loop, including golden sets, trajectory evaluation, LLM-as-judge, and error analysis of production call traces.
  • Use evaluation as a release gate and act on its findings.
  • Improve containment and resolution rates without sacrificing accuracy, and instrument the system to measure outcomes.
  • Work on the voice path, including turn-taking, interruption, latency budgets, and the telephony layer.
  • Participate in carrier calls and listen to recordings to identify and address problems in conversations.
  • Contribute product judgment and insurance context when shaping the agent's capabilities and deciding when to ship an MVP or slow down for accuracy.

Benefits

  • Health insurance
  • Dental insurance
  • Vision insurance
  • Paid time off
  • Company-sponsored retirement plan
  • Some roles may be eligible for an annual company bonus plan, commissions, and/or long-term incentive awards
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service