Insurance-native · In production

The AI back office for
insurance.

Four insurance-native Analysts for MGAs, carriers and reinsurers: they apply the cash, reconcile the bordereaux, structure the submissions, open the claims, and book the cessions. Evidence rides on every row; your PAS stays the system of record.

Reconciliation Analyst
Read bank statement · June48 lines
Read remittances and bordereau3 documents
Matched to policy and binder46 of 48 matched
Queued for review2 items
MTC-0147 Analyst reasoning GBP 12,480.00

Remitter alias MRLD BRKNG LDN resolved to Marmalade Broking, confirmed 12 May. Amount equals net premium less 22.5% commission per broker statement row 41. Reference 7741 appears on the June bordereau, line 18. Matched, queued for approval.

bank line 23 broker stmt row 41 June BDX line 18
46 of 48 matched audit trail: exportable
The Analysts

One Analyst per job. Start where it hurts.

Proof

Proof from production.

80%
“The beauty of Brisc was taking about 80% of the manual labour away, so people can focus on underwriting instead of data entry.”
Nik LuckingManaging Partner, Helix Underwriting Partners. In production across submissions and claims.
CFO
“Brisc has fully automated our ingestion of policy submissions and claims notices, and we are now completing the deployment of our premium receipt reconciliation process.”
Kevin CopelandChief Financial Officer, Helix Global
97%+

Output accuracy across the Analysts in production, with every match carrying the evidence that proves it.

The standard every Analyst is held toNumbers you can put in front of an auditor
Human in the loop

Work the queue.

  • Routine volume clears itself. Uncertain rows queue for review.
  • Every decision shows its work. Reasoning and source documents attached.
  • Never a guess. Matching runs on rules, tolerances and confidence tiers; AI argues borderline cases, never decides a match.
  • Steady state: roughly one row in twenty.
ExceptionRemitterConfidenceStatusAmount
EXC-0341Marmalade Broking94%ReasonedGBP 18,240.00
EXC-0342Lockton Companies88%Needs reviewUSD 486,200.00
EXC-0343Unknown remitter61%Needs reviewUSD 12,850.00
MTC-4118Aon99%ApprovedUSD 3,820,400.00
Autonomy

Trust is earned in stages.

Most teams walk the three stages in about 90 days. You set the pace; the Analyst earns each step.

1Stage one
Match this month’s bordereau

Every match waits for your approval. Your team rejects what is wrong and the Analyst learns your aliases and rules.

2Stage two
Auto-match routine volume, queue the rest

Routine volume clears itself and the match rate climbs from 80% on day one toward 92 to 95%.

3Stage three
Auto-post within tolerance; escalate variances

Matched cash posts to your PAS and GL with the evidence trail attached; exceptions still route to your team.

Deployment

Lands beside your stack.

No re-platforming, no migration, no clean-up project first. Brokers and banks make your data messy; the Analyst reads it as it arrives.

STEP 01

Day one: uploads in, evidenced CSV out

Bank statements, bordereaux, broker statements, exactly as your team receives them. No PAS integration required to start.

STEP 02

Parallel-run a live month

The Analyst runs beside your current process. Your team reviews every match and watches the rate climb.

STEP 03

Auto-push when the numbers have earned it

Matched cash posts back to your PAS and GL automatically, evidence attached. Live in 2 to 6 weeks.

Fit

Keep your PAS. Keep VIPR. Add the evidence layer.

VIPR tells you what was bound. Brisc tells you what was paid.

Trust SOC 2 Type II · GDPR and CCPA compliant · a dedicated Azure tenant per customer, with dedicated database, storage and key vault and no shared fallback · Entra SSO · your data never trains shared models. Security documentation →
FAQ

Evaluating Brisc, answered

Is the matching based on rules or AI, and what does it learn from our corrections?

Matching runs on rules, tolerances and confidence tiers; AI reads the documents and argues the borderline cases, never the arithmetic. Your corrections become three things you can inspect: new deterministic rules, sharper prompts tuned on your data, and a permanent store of confirmed payer aliases. 80% on day one, 92 to 95% by about 90 days.

Which Analyst do I need?

Start from the job. Matching cash to policies: the Reconciliation Analyst. Chasing broker and cedant files into a usable schema: the Bordereaux Analyst. Keying broker submissions: the Submissions Analyst. Opening claims records and clearing laundry lists: the Claims Analyst. Translating cedant statements into cession entries by hand: the Bordereaux Analyst.

Does Brisc replace our policy administration system?

No. Your PAS stays the system of record. The Analyst does the matching and posts one thing back: matched cash against the right premium, with a full evidence trail for why. It isn't a second ledger you have to keep in sync.

How is this different from VIPR or AutoRek?

VIPR tells you what was bound. Brisc tells you what was paid. VIPR structures the bordereau; Brisc validates it against the binder, reconciles it to the cash, and shows the evidence for every match. Most customers run both.

How long does deployment take?

2–6 weeks from kickoff to live. Day one needs no PAS integration: uploads in, evidenced CSV out. Integration comes after the work is trusted.

How is our data handled?

Every customer runs in an isolated tenant. Every match stores its evidence and the full audit trail is exportable. Your data never trains our models. Documentation lives at trust.brisc.ai.

Who is Brisc, and are you backed?

A second-time founding team, self-funded on purpose. Sanjay Malhotra bootstrapped his last company and sold it to Amdocs; Brisc runs on that exit, with no PE or VC on the cap table, so the roadmap answers to customers. In production at Helix Underwriting Partners, SOC 2 Type II, on Microsoft Azure. Ask for the reference call.

What makes Brisc different from generic AI tools?

Three things. The Analysts are insurance-native, built for bordereaux, broker statements, cedant statements, and Lloyd’s market formats. Every output shows its work: matches carry evidence, exceptions carry explanations. And matching is deterministic and rule-based, so the numbers are never a guess.

The walkthrough

Bring your own bordereau. Leave with it matched.

30 minutes, no slide deck, your real files: a bordereau, a statement, a submission. Your business is specific; the walkthrough should be too.

Book the walkthrough