hyperexponential

Novacore is bringing all 20 of its specialty programs onto hx. Admitted, non-admitted, and reinsurance on one platform. Novacore is an independent specialty MGA built on NSM Insurance Group's 35-year track record, working with more than 20,000 agents nationwide. As Novacore CEO Aaron Miller says about the collaboration: "hyperexponential gives us one foundation across admitted, non-admitted, and reinsurance, so we can spot trends early, respond deliberately, and stay ahead of the market for our agents and carriers." Read the full release: https://lnkd.in/eu78aCRj

Agents can now do the heavy lifting around an underwriting decision using a carrier’s own pricing logic, workflows, portfolio context, and governance. That was the step change hx unlocked this year. Years of building the underlying platform made it possible. Now, a rating algorithm built in hx is a tool an agent can call. Underwriting logic captured in hx is organizational memory an agent can act on. And every agent action runs inside the carrier’s authority model, permissions, review points, and audit trail. As Amrit Santhirasenan puts it: “Our platform is now the source of the tools and context.” That is how agent-led underwriting becomes operational. From hx Live London.

Amrit Santhirasenan, Co-Founder and CEO of hyperexponential, on building AI for deep vertical markets: “There are a small number of things that professionals do to a very high degree of craft. It has always been our job to respect and honor that.” The goal isn’t broad capability. It’s working backwards from specific professional jobs and building the depth those jobs demand. Horizontal tools can aim to be 80% good at 100% of things. hx needs to be 100% good at the highly specialized work the best underwriting decisions depend on. From Amrit’s conversation with Joe Schmidt at a16z.

"Instead of the underwriter saying, 'I need to make this decision,' the agents are going to tell the underwriter, 'You need to make this decision.'" Karlyn Carnahan, Head of Insurance, North America at Celent, on what’s next after AI chatbots and data ingestion pilots, and why carriers need to map the underwriting work that never lived in the policy admin system. She’ll be sharing more from this research at hx Live New York, on September 17th! Request your invite at https://lnkd.in/eQapuwJg//?utm_source=linkedin&utm_medium=paid&utm_id=893704474&utm_campaign=HX-ThoughtLeadership-AllObjectives-SC&hsa_acc=509625713&hsa_cam=893704474&hsa_grp=879403314&hsa_ad=1504386244&hsa_net=linkedin&hsa_ver=3

Non-agentic underwriting workbenches rely on underwriters to do the vast majority of the work. Intake and triage may be automated. Pricing logic, risk assessment, and portfolio context usually aren't. And the workbench is still a browser window you have to go to. With hx's agentic underwriting workbench, hyperoperator does the heavy lifting and meets underwriters where they are: in their email inbox, Teams channels, mobile notifications. Come see it in New York, at hx Live, September 17. Request an invite using link in the comments below.

Most underwriting AI never leaves the first 20% of the workflow. Over the last two years, ingestion, triage, and routing got faster. The 80% that drives profitability — pricing, risk assessment, portfolio steering — mostly hasn’t changed with AI. Until now. Amrit Santhirasenan, our Co-Founder & CEO, on why solving the easy part first isn’t the same as solving the part that matters. Read the full blog here: https://lnkd.in/e_-ZXjjb//?utm_source=linkedin&utm_medium=paid&utm_id=881503274&utm_campaign=HX-TrafficCampaigns-AllObjectives-SC&hsa_acc=509625713&hsa_cam=881503274&hsa_grp=841412954&hsa_ad=1486786394&hsa_net=linkedin&hsa_ver=3

Amrit Santhirasenan, our Co-Founder and CEO, on the core underpinnings of an agentic underwriting system (and why an AI model alone isn't enough): "There are three pillars upon which you build a system like this: the tools, the context, and the environment. The tools are the things that agents use to do the work. The context is the business logic it uses. And the environment is the infrastructure and controls that operate and constrain the system." "Our platform is now the source of the tools and context. When you design and build a rating algorithm, it becomes available as a tool for the agents to work with. When you write your underwriting logic down and institutionalize it in the system, that becomes available too. And then all of this becomes context, and it means the agent can do the heavy lifting for you." "When we got the system up and running, I was wowed at what was possible. But it's actually everything we've been building before that has led to this moment." Missed Amrit's talk at hx Live London? hx Live returns on September 17 with hx Live New York. Request an invite using the link in comments.

Liverpool Street Station sits at the heart of London’s insurance market, where hx got its start. We could think of no better place to celebrate last week's launch of hyperoperator (hyperoperator.ai). hyperoperator isn’t generic AI retrofitted for insurance. It’s the first real agent for end-to-end underwriting work, built by the market, for the market.

"If your pricing model is built in Excel, every time you're releasing a model, that means there's a new version of the Excel model floating around that people might end up using." Joel Smith, DUAL's Head of Pricing, on what model releases used to look like before DUAL rebuilt its financial lines pricing on hx. Read the full DUAL story: https://lnkd.in/eRpXv7Qg//?utm_source=linkedin&utm_medium=paid&utm_id=881503274&utm_campaign=HX-TrafficCampaigns-AllObjectives-SC&hsa_acc=509625713&hsa_cam=881503274&hsa_grp=841412954&hsa_ad=1486786384&hsa_net=linkedin&hsa_ver=3