Guidewire launched the Qusar release with an Agentic Framework that lets insurers build and manage AI agents directly inside the Guidewire Cloud Platform. The framework gives agents real-time access to policy, claims, and billing data, turning multi-step processes that used to take days into minutes.
"Insurers are increasingly recognising that AI value comes from deep integration with core business processes, not from isolated experimentation," said Karlyn Carnahan, executive partner at Celent. "Agentic Framework addresses this directly by letting insurers deploy AI within their existing systems and workflows while maintaining operational control and compliance."
Built-in insurance agents
Qusar ships three Guidewire-developed agents for specific workflows. Claim Summarisation for ProNavigator reads adjusters' notes and creates concise summaries so adjusters spend less time reviewing text and more time resolving complex claims. Policy Change for ProNavigator acts as an embedded assistant for underwriters and service reps, helping them complete policy changes faster to improve quote-to-bind ratios. Agentic First Notice of Loss (FNOL) uses conversational AI voice to guide claimants through the reporting process and capture key details.
"Guidewire AI innovations are helping us transform how we support our teams and serve our members," said Garrett Anderson, CIO of the Automobile Club of Southern California. "By using AI to summarise claims and streamline key insurance workflows, we're improving efficiency, helping our adjusters focus on higher-value work, and creating a stronger foundation for future innovation."
Developer assistants that understand Guidewire code
The release also includes Developer and Builder Assistants trained on Guidewire's code base, programming languages (including Gosu), design patterns, and configurations. These assistants help developers ship features more than 40% faster than generic coding assistants, according to Guidewire. For data and product teams, a Data Curation Assistant for Data Studio turns plain language into precise SQL, and a Product Design Assistant for Advanced Product Designer automates insurance product configuration.
"By building the Agentic Framework directly into Guidewire Cloud Platform, we are giving developers and AI builders the tools to engineer and safely deploy insurance-aware AI agents into their daily operations," said Diego Devalle, chief product development officer at Guidewire.
For product development teams looking to apply similar AI Agents & Automation techniques in their own stack, the deep integration with existing systems is the key takeaway. Guidewire's approach avoids bolt-on chatbots and instead embeds agents directly into claims, underwriting, and policy workflows - a model that any organization building internal AI tools can study. The focus on AI for Insurance also shows how industry-specific training, rather than generic models, produces assistants that actually speed up real work.
Why this matters for product development professionals
Guidewire is showing that the real AI value for insurance comes from embedding agents into existing business logic, not from isolated experiments. For product teams, the 40% speed improvement on feature delivery - measured against generic coding assistants - suggests that domain-specific fine-tuning can produce far better developer productivity gains than off-the-shelf tools. If you build internal tools for niche workflows, this release is a concrete case for investing in training on your own code and domain models rather than relying solely on general-purpose AI.
Your membership also unlocks:
