AI & ML interests

AI & ML for Actuarial Science

Recent Activity

Organization Card

Intelligent Actuaries

An SI lab for actuarial science.

GoogleLogoGIF-LARGE

We research and build SI systems for the actuarial function: agents, models and methodology, evaluated against the regulations our partners file under. By introduction. Under mandate.

This Hugging Face organisation is the public face of the lab's research output: preprints, reproducible simulators, methodology notes. The institutional workbench itself is private.

Website Engagement Research Lab


Research themes

The lab publishes under three themes. New work accumulates beneath each.

  • Nanoeconomics: Community-scale wealth modelling under cooperation. The W(M, T, R) framework, its empirical pillar (Community Rental Survey 2022), and its stochastic survival simulation.
  • Actuarial methodology: Reserving (chain-ladder, Mack, bootstrap), mortality (Lee-Carter, CBD), capital (ESG paths, SAM SCR), researched in the open, deployed under mandate.
  • Climate & catastrophe: Re-analyses, EP curves, parametric triggers, hazard layered onto exposure with provenance to source. Work in scope for 2026.

Currently available

  • Nanoeconomics: Compounding Cooperation for Community Wealth, extended preprint with a fully reproducible stochastic survival simulation of W(M, T, R). Read.
  • W(M, T, R) survival simulator, single-community Monte Carlo and society-network modes, running locally in the browser. Open.
  • SoftData workstation, the intake desk for raw datasets: column-type inference under audit, op-by-op cleaning, reproducible exports in Python, R, C++ and SQL. Open.
  • Synthetic Term & Whole Life Lapse and Surrender Study 2023–2025, 10,000 synthetic policyholders answering the Society of Actuaries (SOA) Research Institute and LIMRA 2023–2025 lapse and surrender experience study data request, as 27,692 VM-51 policy-year records. Dataset.
  • Lapse & Surrender Study · scelo, the same study analysed live in a Gradio app built on our scelo Python library: experience rates, the lapse GLM, and lapse-adjusted pricing, IFRS 17 CSM and Solvency II SCR. Open.

Acknowledgements

Our lapse and surrender study answers the Society of Actuaries (SOA) Research Institute and LIMRA 2023–2025 Term and Whole Life Lapse and Surrender Experience Study data request, which defines its scope and its VM-51 record layout. We gratefully acknowledge the SOA Research Institute and LIMRA for that request and for the experience studies it builds on. The data is synthetic, and the work is independent: it is not SOA or LIMRA data, and it is not affiliated with or endorsed by either organisation.


How we work

Research, in the open. Preprints, methodology notes, reproducible simulators. Methodology travels with every run: seeds, parameters, citations exposed in the UI.

Deployment, by mandate. No public binaries. No public pricing. Engagements with tier-1 insurers, reinsurers, sovereign funds and pension boards, deployed inside the client perimeter under a Master Engagement Agreement.

Every figure traces to its input, the model that produced it, and the regulation it sits under. Defensible to the board. Auditable to the regulator.


Engagement

We accept a small number of new counterparties each year. Fees quoted privately, per mandate, after introduction.


Where to find us


Filed under FSCA · Prudential Authority · IFRS 17 · SAM · POPIA.

Built in South Africa 🇿🇦.

models 0

None public yet