Why I’m building Bindflow.

I’m not an insurance person. That matters — I’ll get to why.

My background is government and public infrastructure. Prior to Bindflow I spent years building things inside and around the Government of Canada — the kind of work where nobody knows your name but thousands of people depend on the pipes you’re building being reliable. I learned two things from that period. First, most institutions run on invisible systems that are far more fragile than anyone realizes. Second, the people actually using those systems — the CSRs, the clerks, the operators — are doing heroic work holding it all together with plug-ins, duct tape, and shared spreadsheets.

Commercial insurance brokerage is that same world.

When I started looking at the space in early 2026, what I found was an industry where the most expensive labor in the chain — your CSRs, your producers — spends most of their day re-keying data that was already typed once upstream. Every independent commercial agency in America runs some variation of the same Tuesday morning: a producer forwards an ACORD, a CSR retypes 140 fields into an AMS, she guesses at six carriers, she drafts six slightly-different emails, she waits. Three of the six were never going to quote. Two come back asking for attachments she forgot. By Friday the account is gone to a faster agency across the street.

There’s nothing about that process that couldn’t be fixed ten years ago with ordinary software. It wasn’t, because insurance incumbents built tools for carriers and enterprise brokers, not for the 40,000 independent agencies that write most of America’s small- and middle-market commercial premium. LLMs change the math — ACORD extraction, carrier appetite matching, and submission drafting are finally cheap enough that a small team can ship a real product for a small-agency buyer. That’s the window I’m building into.

Why a non-insurance generalist building in insurance: because the industry has been iterating on the same assumptions for thirty years. The people who know commercial P&C best are also the people least positioned to ship software fast enough to catch this cycle. Incumbents like Applied and Vertafore have 3-year product backlogs. They’re not going to build what independent agencies need quickly — the broker-side of insurance needs someone outside the industry, shipping in weeks, paying close attention to the actual work of the CSR who’s going to use this thing every day.

That’s me. If you’re a commercial principal or ops leader reading this, I’m the person you’ll talk to on every sales call. I wrote the code. I’ll implement the integrations. I’ll answer your Slack messages at 10pm. For the first 10 pilot agencies, you get the founder, not a support tier.

My commitments to the first 10 pilot agencies:

• Every integration we ship for your shop during the pilot is yours to keep if you don’t convert.
• If we don’t move your numbers by Day 90, I refund the pilot in full.
• If we do move your numbers and you convert, you’re locked in at pilot-cohort pricing for 24 months, even if we raise rates for later customers.
• You co-author the case study with us. We pay for the design. You get final edit.

That’s the pact. If it sounds reasonable, I’d like 20 minutes of your time.

— Tom Hwang, founder, Bindflow. Ottawa, Canada. · @TomSungWooHwang