Qlo, a Y Combinator Summer 2026 company, has launched AI agents that carry out commercial insurance underwriting work, and says an enterprise customer tripled its weekly primary quote throughput after going live in May 2026. The figure is company-reported, and no independent audit has verified it.
The startup is entering a crowded field. Insurers and managing general agents, the specialist firms that underwrite on an insurer's behalf, are pushing more of that work onto software agents, and hyperexponential, Federato and Cytora have already claimed similar ground.
What Qlo is selling
According to founder Ryan Hu's YC launch announcement, Qlo's agents execute underwriting instructions. The company says any future autonomy is meant to stay inside each insurer's appetite and controls.
Qlo's LinkedIn page lays out its strategy in three parts: automate the operational work surrounding underwriting decisions, build a shared system of record, and widen the agents' autonomy over time. Humans and agents would work from the same accounts, documents, permissions, rules and workflow state. The fully autonomous underwriter is where the company says it is headed. It is not a description of what runs in production today.
Hu acknowledged as much in an undated LinkedIn post, saying the full platform is still being built. He has written separately about guardrails and human control. Qlo's website describes shared records and auditable actions, but those are the vendor's own characterizations and have not been independently tested.
The customer result
The YC announcement said an enterprise customer signed a two-year contract, went live in May and tripled weekly primary quote throughput. In a Tella demo video indexed July 7, 2026, Hu said production began in early May.
He credited the gain to submissions that had previously sat untouched in the inbox, according to an undated LinkedIn post.
Walnut Insurance, in an August 23, 2026 overview of YC's summer batch, identified the customer as Distinguished Programs Executive Lines and named Ryan Becker as its president. Qlo's website carries a testimonial from Becker, whose company is styled there as Distinguished Executive Lines; the naming difference is unexplained. The testimonial is undated and hosted by the vendor. Walnut's supporting links lead back to Qlo and YC, so its coverage cannot be treated as a separate check. As of November 8, 2026, Qlo's LinkedIn feed was still repeating the threefold claim.
Pricing and onboarding
Qlo says it onboards programs using their existing forms and rates and leaves their policy administration system in place. It advertises implementation within two weeks, which is an offer rather than a measured average.
Walnut reported a 30-day, fully refundable paid pilot, followed by a flat first-year fee within a two-year partnership. Qlo's current site gives no comparable public pricing, so those terms may have changed.
Hu has also promoted a bordereaux reporting feature, the periodic premium and claims reports that program administrators send to their insurers. In a LinkedIn post he tied it to future portfolio insights, rating-model changes and analysis of underwriting performance. No published outcome data supports those benefits yet.
A small team, a standard deal
Qlo's YC profile lists two people. LinkedIn shows a 2-10 employee band and four associated employees. The two counts measure different things, and exact payroll headcount remains unverified. The company lists Taipei and San Francisco as locations.
Hu studied electrical engineering at National Taiwan University and left before finishing, according to his YC biography. He moved from Taiwan to build the company in the United States.
Qlo has announced no financing round. Dealroom lists $125,000 from Y Combinator in June 2026, labeled seed. YC's published standard deal is $500,000, made up of $125,000 for 7% and $375,000 on an uncapped MFN SAFE, a convertible instrument that gives the investor the best terms offered to later backers. Qlo's executed documents could not be located, so it is unconfirmed that its terms match.
Rivals are moving fast
On July 8, 2026, hyperexponential announced hyperoperator, a system that coordinates specialist agents from broker email through pricing and quoting, within carrier appetite and authority controls. It said a controlled rollout would follow later in 2026. On October 27, 2026, Guidewire signed an agreement to acquire hyperexponential for $1.7 billion in cash, plus up to $400 million in contingent consideration. Whether the deal has closed is not established in the material reviewed.
Federato described end-to-end agentic underwriting with auditable citations on May 12, 2026. Cytora announced its Autopilot product for release on March 30, 2026, and on September 22 reported a Zurich deployment spanning more than 20 countries with an 80% cut in manual triage time. Those numbers come from vendors and their customers, and they are not comparable with Qlo's throughput measure.
Governed agents, underwriting context and workflow orchestration already feature in competitors' own marketing. On the evidence available, Qlo's architectural pitch is not unique.
The market behind the race
Conning estimated on July 28, 2026 that U.S. MGAs wrote about $128 billion in premium in 2025. Statutory filings captured $102.6 billion of that, up 12% from 2024. AM Best reported on June 30, 2026 that delegated underwriting authority enterprises sourced $108.7 billion in direct premiums in 2025, compared with $92.3 billion the year before. The two sets of figures cover different scopes and should not be read as competing estimates of the same thing.
Qlo also presented at InsurTech NY's Match Demo Day, and the organizer's LinkedIn account commented on its participation. That confirms a place in the program. It does not show an investment or a customer contract.
Hu's own yardstick remains the inbox, and the pile of untouched submissions that started it. The company's next test is whether the threefold result holds at a second customer.
