AI for reinsurers and TPAs, for the document work that scales badly with headcount.
Bluegrass builds production AI systems for reinsurers, third-party administrators and the service firms around them, from bordereaux normalization to exposure roll-ups. They run in your own cloud, load clean data into your warehouse and belong to you.
These businesses sit downstream of everyone else's formats, so data arrives in whatever shape the cedent, MGA or client chose. Making it usable is a cost paid again every month.
Where the time goes
Inbound bordereaux
Premium and claims bordereaux arrive from many sources, each with its own columns, codes and quirks.
Schedules and SOVs
Statements of values and exposure schedules arrive as spreadsheets and PDFs that have to be cleaned before they can be modeled.
Exposure roll-ups
Portfolio views depend on combining all of that data accurately, and a roll-up is only as good as its worst source.
Claims and service files
TPAs read and act on large volumes of claim files, correspondence and reports, and the reading grows with every new client.
What we build for reinsurers, TPAs and services firms
Schedule extraction
Statements of values and exposure schedules mapped from any layout onto one schema, with addresses standardized and gaps flagged.
Exposure roll-ups
Validated data combined into portfolio views by cedent, program, peril and territory, traceable back to the source files.
Claims file summarization
Claim files summarized and kept current as documents arrive, so that handlers and clients can see the position without rereading the file.
These are examples of what we can build for reinsurers, TPAs and services firms. Pick one up as-is, have it tailored to you or suggest something entirely new.
Talk to an engineerWhere to start
We usually recommend starting with normalization, getting inbound data into one validated schema, because every downstream analysis and report depends on it.
The 4-week Opportunity Evaluation measures how much of your inbound data can be mapped automatically, and what that is worth, before anything is built.
Questions and answers for reinsurers, TPAs and services firms
- Can the data load into our own warehouse?
- Normalized data is written to your own warehouse in your own schema, so your analysts and models keep working with the tools they already use.
- How do you handle a new cedent or MGA?
- A new source's format is mapped once and confirmed by a person, then handled automatically, so onboarding a new source becomes a review rather than a project.
A senior team that has done this before
We are a small, entirely senior team with fifty years of combined AI engineering, multiple granted patents and Y Combinator founders, from Yale, MIT, McKinsey, Goldman Sachs, Microsoft and Vista Equity Partners. The people who scope your work are the people who write the code.
Selected work
We built the AI transformation roadmap for a top-50 P&C brokerage, and it is saving the firm millions of dollars in vendor fees it did not need to pay.
Clients include some of the largest banks, RIAs, fintechs and AI labs, as well as a top-3 P&C carrier, a top-10 P&C brokerage, a top-5 trucking brokerage, a leading specialty MGA and a leading trucking MGA.
We run only 3 to 5 engagements at a time, so that every client works directly with the senior team.
About the team- Dozens
- of insurance companies, RIAs, banks and fintechs we have worked with
- 50 yrs
- combined AI engineering experience
- 10+ Patents
- granted to members of the team
- 100%
- of the IP we build is assigned to you
Tell us which workflow costs your team the most time.
A 30-minute call with an engineer usually shows whether something is worth building. Within a week you will have a free written point of view on the two or three opportunities we think are real and what each is worth.