info@marzallabs.ai
Clinical team standing together in a hospital

NHS & Public Sector

Clinical AI that ranks urgency
without replacing judgement

DSPT-compliant deployment on NHS-approved infrastructure. Discovery to production in twelve weeks, including information-governance sign-off.

  • NHS acute trusts
  • Referral management
  • RTT pathways
  • DSPT
  • CQC
  • Public sector procurement

Referral volumes keep rising against an 18-week RTT pathway target that does not move, and triage is still largely done by clinicians reading unstructured referral letters of highly variable quality. Every touchpoint carries DSPT, UK GDPR and CQC obligations. Procurement timelines and information-governance sign-off are real constraints rather than paperwork, and any system that ignores them does not get deployed no matter how well it performs.

We design inside those constraints from the first architecture session. Deployment runs on NHS-approved infrastructure with patient data never crossing the approved boundary, human-in-the-loop review is mandatory rather than optional, and the audit trail is built to CQC expectations. At one acute trust that meant twelve weeks from discovery to production — including IG sign-off, not excluding it.

Capabilities

What we build for nhs and public sector

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

Referral triage and prioritisation

Referral letters read, structured clinical data extracted, and clinical urgency scored across pathways.

  • 2,400 referrals per month triaged automatically
  • 7 specialty pathways scored, each with XAI reasoning
  • 62% reduction in admin time per referral

Mandatory human-in-the-loop

A clinician reviews every score before anything routes. The system ranks; the clinician decides.

  • 100% of scores clinician-reviewed before routing
  • Automatic escalation on urgent clinical indicators
  • Every override captured and retained with the record

Information governance by design

DSPT-compliant deployment with a Data Processing Agreement and patient data held inside the approved boundary.

  • 100% DSPT-compliant — NHS-approved infrastructure only
  • UK GDPR Data Processing Agreement in place before build
  • Data never leaves the NHS-approved boundary

CQC-ready audit trail

Every triage decision explained, attributed and retained in a form an inspection can follow.

  • Full audit log on every triage decision
  • Clinical rationale retained alongside the AI reasoning
  • Structured for CQC evidence requirements

Document and correspondence processing

Extraction and classification across referral letters, discharge summaries and administrative correspondence.

  • Handles unstructured letters of variable quality and completeness
  • Missing-data flags raised explicitly rather than inferred
  • Structured output into existing trust systems

Delivery inside procurement reality

Programmes scoped and sequenced around NHS procurement and IG timelines rather than despite them.

  • 12 weeks discovery to production, including IG sign-off
  • 0 RTT breaches in the three months after deployment, from 14 the prior quarter
  • Knowledge transfer to trust staff built into the engagement

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

Shorten the pathway without moving a single clinical decision away from a clinician.

Talk now