Actuarial Report Drafting Agent

The Actuarial Report Drafting Agent takes this year's model outputs and last year's report and drafts the statutory report around them. It writes the narrative, explains the analysis of change, builds the reconciliations and carries forward every disclosure the report needs, for insurer actuarial function reports, retirement fund valuations and medical scheme pricing reviews. Every number is tied back to the model output it came from, so actuaries spend their time on judgement, not on rewriting last year's words.

Ideal forConsulting actuaries

The difference

From 3-5 days of drafting per report to 3-4 hours

Your actuaries sign off more reports in the busy season, each one consistent with the model and with last year.

Done by hand

3-5 days of drafting per report

With the Vanine agent

3-4 hours

Time returned to your team

85%

Indicative figures. We measure your own baseline in the first fortnight so the numbers you see are yours.

Actuaries spend time on judgement

The drafting, roll-forward and tie-outs are done, so the actuary focuses on assumptions, results and advice.

Numbers that tie, every time

Every figure in the narrative is reconciled to the model output, so review notes about mismatched numbers disappear.

More reports through the busy season

Valuation and pricing deadlines cluster, and the same team can deliver more of them on time.

Capabilities

The narrative, analysis of change and reconciliations drafted from your model outputs, before the actuary starts writing.

Built for consulting actuaries, valuators and heads of actuarial function at South African insurers, retirement funds and medical schemes.

Report roll-forward

Carries last year's report into this year's structure and updates every date, figure and reference.

Analysis of change commentary

Explains the movement in liabilities, surplus or contributions step by step, from the model's analysis of change.

Reconciliations

Builds the reconciliations between model outputs, financial statements and the prior valuation, and flags any that do not close.

Assumption and method disclosure

Drafts the assumptions, methods and data sections, and highlights what changed from last year.

Regulatory checklist

Checks the draft against the required content for the report type, from the insurer actuarial function report to the statutory retirement fund valuation.

Tied to the model

Links every figure in the narrative to the model output or schedule it came from.

How it works

From your systems to a result you sign off

Your actuary reviews, edits and signs every report.

TeamsExcelWordSharePointOutlook
Any system you use
Connects

Connect the places your actuarial valuations already live.

Model outputs, valuation data, last year's report and the financial statements from your shared drives and modelling environment. Already use something else? We connect that too.

  • OpenAI Frontier
  • Your own AI platform or models
  • Microsoft 365
  • Actuarial modelling outputs (Prophet, R, Python and Excel models)
  • Prior-year reports and working papers
  • Word report templates and Excel reconciliation schedules
Works in

Runs when the model run is signed off, or on request in Teams, using the models your organisation approves.

Report roll-forward
Analysis of change commentary
Reconciliations
Assumption and method disclosure
Regulatory checklist
Tied to the model
Delivers to

A draft report in your Word template and a reconciliation workbook in Excel, saved to the engagement folder.

Step by step

  1. 01

    The model run for the valuation or review is signed off.

  2. 02

    The agent reads the outputs and last year's report.

  3. 03

    It rolls the report forward and updates every figure.

  4. 04

    It drafts the analysis of change and the reconciliations.

  5. 05

    It checks the draft against the required content and flags gaps.

  6. 06

    Your actuary reviews, edits and signs the report.

Deployment

Runs inside the tools your team already uses

The same agent and the same cited output, delivered three ways: through the AI assistant you have rolled out, on a model you host yourself, or inside your own software.

Model Context Protocol

In the frontier assistant you already pay for

We publish the agent as an MCP server. Add it as a connector and your team calls it from the chat they use every day, with no new tool to learn.

Adopting frontier AI across your teams
Called from ClaudeMCP connected

"Roll last year's valuation report forward"

Toolactuarial_report_drafting.roll_forward_report
report
Statutory valuation 2026
fund
Pension fund, R2.4bn
Returned42 sections updatedEvery finding linked to its source document
Self-hosted

On a model you host yourself

Run the agent on open-weight models in your data centre or private cloud. Your actuarial valuations never leave your network, which keeps POPIA and data residency straightforward.

  • Llama
  • Mistral
  • Ollama
  • or any OpenAI-compatible endpoint
Sovereign AI on your infrastructure
Bespoke systems

Inside your own software

Trigger a run from your own workflows, and push the finished output into your ERP, GRC or reporting systems. We build the integration with your team.

  • REST API
  • Webhooks
  • Batch jobs
  • Embeddable review panel
Custom integrations from our software factory
Governance

Governed the same way, every route

Whichever way your team reaches the agent, identity, permissions and audit stay with your existing controls.

  • Single sign-on through Entra ID, Okta or Google Workspace
  • Reads only the documents each user is already allowed to open
  • Every tool call written to an audit log you can export
How your data is handled

DATA SECURITY & PRIVACY

Your actuarial valuations stay yours.

The agent runs inside your tenancy, reads only what each task needs, and never trains on your actuarial valuations. Every step is logged so your auditors can see exactly what it touched.

Built for regulated work.

+
WHAT THE AGENT KEEPSONLY THE OUTPUTENCRYPTED AT REST0 BIT AES0% DELETABLE0 TRAINING RUNS

YOUR ACTUARIAL VALUATIONS NEVER LEAVE YOUR ENVIRONMENT

Precision AI for Institutional Workflows

Build once.Deploy across teams.Improve over time.