info@marzallabs.ai
The City of London skyline in daylight

Specialty Insurance

Reimagine underwriting with
AI built for the London market

Slip-literate. SOV-literate. MRC v3-aware. Three products live in production inside specialty insurance businesses today.

  • Lloyd’s syndicates
  • MGAs
  • Coverholders
  • Specialty carriers
  • US E&S
  • Delegated authority

Rate adequacy is under pressure at the same time as expense ratios are being scrutinised line by line. Blueprint Two is changing how data moves through the market, the FCA’s expectations on AI-assisted decisions keep tightening, and delegated authority oversight has become a board-level topic. Meanwhile the work itself has not changed: a submission still arrives as a PDF and an email, a renewal still means comparing two slips clause by clause, and a schedule of values still lists four hundred assets in a dozen formats.

We build for that reality rather than around it. Our products read the documents specialty insurance actually runs on — slips, SOVs, wordings, loss narratives — and every output carries an explanation traced back to the source, because a decision an underwriter cannot defend is a decision they will not use. Nothing binds, nothing prices, and nothing routes without a human signature where one is required.

Capabilities

What we build for specialty insurance

Each capability below is delivered through named assets that already exist. Figures are delivered results unless marked otherwise.

Submission intake and triage

Every inbound submission read, screened, scored against appetite and routed with a context pack attached.

  • 4.2 days to 18 minutes end to end, in production
  • 6 underwriting criteria scored per submission, each XAI-traced
  • 400+ submissions per week handled at one MGA

Renewal and wording comparison

Baseline-versus-renewal slip review as a structured analysis with severity attached to every finding.

  • 52 sections analysed in a median two minutes
  • 11 classes of business, D&O through offshore energy
  • Financial deltas on premium, limits, retentions and sub-limits

Exposure and asset classification

Assets classified from schedules, slips, SOVs and loss narratives, with your firm’s mappings retained.

  • 97.4% accuracy on a live instance
  • Under 60 seconds per classification including reasoning
  • Every human correction captured and applied to the next case

D&O and financial lines research

Eight underwriting risk vectors assessed from primary sources, every finding cited.

  • 5 minutes per company, from roughly four hours
  • 8 risk vectors assessed on every submission, none skipped
  • AI liability, ESG fiduciary risk and ECCTA covered as standard

Claims triage and loss analytics

Classification, routing and structured extraction from loss narratives and claims correspondence.

  • Loss narrative extraction into structured, comparable fields
  • Routing rules built on your own claims handling thresholds
  • Full audit log on every automated classification

Underwriting and portfolio analytics

A governed semantic layer with agreed definitions of submission, risk, policy and claim.

  • One agreed definition per term, respected downstream
  • 180+ automated quality tests on every pipeline refresh
  • Delivered on Fabric, Databricks or Snowflake — your estate, your choice

Named assets

The inventory we draw on

Twenty-one products, diagnostics, agent patterns, platform components and governance frameworks. Filter by category.

  • PolicyCompareIQClause-level policy and slip comparison
  • ClassifierIQAsset classification with a firm-specific memory loop
  • D&O AnalyzerIQEight-vector D&O underwriting research
  • TriageIQSubmission intake scoring and routing — 2026
  • AI-Ready BlueprintThree-week fixed-scope data estate diagnostic
  • AI Readiness AssessmentData, technology, capability and governance scoring
  • Data & AI Roadmap DesignSequenced, costed initiative plan with owners
  • AI Operating Model DesignReviewer roles, escalation paths, HITL workflow design
  • Technology Stack AdvisoryPlatform selection against your estate and constraints
  • AI Agent Architecture & DesignTask boundaries, tool permissions, failure modes
  • Document Processing AgentsMulti-format ingestion across email, PDF and ACORD
  • Underwriting & Triage AgentsAppetite scoring, capacity checks, context packs
  • Claims Processing AgentsClassification, routing and loss-narrative extraction
  • RAG Knowledge BasesRetrieval over your wordings, guidance and precedent
  • Modern Data Platform DesignMedallion architecture on Fabric, Databricks or Snowflake
  • Data Governance FrameworkOwnership, lineage and automated quality rules
  • Data Quality Programmes180+ automated tests running on every pipeline refresh
  • Underwriting & Claims AnalyticsSemantic layer and agreed portfolio definitions
  • Databricks Partnership ServicesLakehouse delivery under the Bronze partnership
  • AI Compliance & GovernanceFCA, Lloyd’s, EU AI Act and GDPR in one register
  • ROI & Value FrameworkBaselines, KPIs and attribution set before build

Partnerships

The platforms behind the delivery

Databricks Bronze partner, with production delivery on every platform named here.

  • Microsoft AzureCloud, HITL pipelines, Azure OpenAI
  • Azure AI FoundryModel catalogue and agent tooling
  • DatabricksBronze partner · Delta Lake and ML
  • Google CloudCloud platform and data services
  • Anthropic ClaudeDocument reasoning
  • OpenAIExtraction and analysis
  • Microsoft FabricUnified analytics and OneLake
  • SnowflakeData cloud analytics
  • LangChainAgent orchestration
  • dbtTransformation and testing

Unify submission, renewal and exposure intelligence into one measurable decision system.

Talk now