Rebuilding Underwriting From the Inside Out: The AI OS for Modern MGAs

Introduction

What if the real bottleneck in commercial insurance isn't distribution or pricing—it's the workflow itself? Nearly $100 billion of SME P&C insurance is placed every year using manual processes, disconnected systems, and data that lives in spreadsheets and email threads. Hamesh Chawla has spent the last four years building the infrastructure to change that.

Before founding Mulberri in 2021, Chawla led product and technology at Edelman Financial Engines and Asurion. He came to insurance not as a lifer but as a technologist who saw an industry still running on 20th-century tooling. Mulberri is his answer: an AI operations platform connecting PEOs, brokers, SMEs, and carriers—from smart submission and risk scoring to quote-and-bind and certificate of insurance.

In this conversation, Josh Hollander and Chawla dig into why the MGA market was the right pivot, what AI governance looks like when binding decisions carry real capital risk, and why the SME segment is the most underserved frontier in commercial insurance.

Guest Bio

Hamesh Chawla is the Co-Founder and CEO of Mulberri, an AI operations platform for MGAs, PEOs, brokers, and carriers serving the SME market. Before Mulberri, he was EVP and Chief Product & Technology Officer at Edelman Financial Engines, with prior roles at Asurion and Zephyr (acquired by SmartBear). He holds an MS in Computer Science from Texas A&M University. Mulberri has raised $10.8M from Eos Venture Partners, Altamont Capital Partners, MS&AD Ventures, and Hanover Technology Management.

Key Topics

• The $100B manual workflow problem — Nearly $100B of SME P&C is placed annually using ACORD forms emailed back and forth, loss runs parsed by hand, and decisions made without the data that exists in the market. Mulberri automates this stack.

• From embedded insurance to AI operating system — Chawla explains why he pivoted from embedded distribution to building the workflow layer MGAs actually run on—ingesting unstructured data, structuring it through a GenAI OS, and routing decisions with full context.

• AI governance when capital is at stake — When AI is binding real policies, black-box models get rejected. Mulberri surfaces claim propensity, frequency, severity, and loss ratio so underwriters can interrogate and trust the output.

• The PEO channel as data and distribution — PEOs sit on firmographic and workforce data directly predictive of workers' comp risk. Embedding into that channel is both a data strategy and a go-to-market strategy.

• Building for carriers, brokers, and SMEs simultaneously — Carriers need loss ratio visibility, brokers need submission efficiency, SMEs need straightforward access. Aligning all three is the hardest product problem in the space.

Notable Quotes

"Our mission since day one has been to leverage technology to complement underwriters' expertise—simplifying and streamlining the business insurance process while ensuring transparency."

"The Risk Engine puts the information underwriters need at their fingertips to make fast, accurate decisions—not replacing them, but making them dramatically more effective."

Resources

Guest:

• Mulberri: https://www.mulberri.io

• Hamesh Chawla on LinkedIn: https://www.linkedin.com/in/hameshchawla/

Host & Organization:

• Joshua R. Hollander on LinkedIn: https://www.linkedin.com/in/joshuarhollander/

• Horton International (USA): https://www.horton-usa.com/

• Insurtech Leadership Podcast (LinkedIn Showcase): https://www.linkedin.com/showcase/insurtech-leadership-show

Subscribe & Review

If you enjoyed this episode, subscribe on your favorite platform and leave a review. The Insurtech Leadership Podcast is available on YouTube, Apple Podcasts, and Spotify.