An actuary who turns uncertainty into decisions.
I'm Jonas Osman Abdelghafour — professionally known as Jonas Osman. For more than twenty years I've helped banks, insurers, reinsurers, and pension schemes measure the risks on their balance sheet and make better decisions about them.
My work sits at the intersection of actuarial technique, quantitative finance, and regulation. I build IFRS 9 ECL and credit-risk models that hold up in audit, IRRBB and ALM frameworks that survive supervisory challenge, Solvency II reserving and capital models that inform real business decisions, and model validation opinions that regulators actually accept.
Increasingly, that work extends into emerging risks — climate and catastrophe modelling, sovereign and geopolitical risk, and the careful application of machine learning and AI inside risk functions where explainability and governance matter as much as accuracy.
Clients typically bring me in when the stakes are high and the answer isn't obvious: an IFRS 9 model that has to be re-calibrated in weeks, an IRRBB framework a regulator has challenged, a capital model that needs to be defensible and useful, or an AI model risk framework built to survive both internal governance and the EU AI Act.
I work independently through Quantica Risk Model. No large project team, no hand-offs to junior consultants — you get the person you hired.
Areas of expertise
- Enterprise Risk Management
- Actuarial Science
- Loss Reserving
- Capital Modeling
- Predictive Modeling
- IFRS 9 & Credit Risk
- Solvency II & ORSA
- IRRBB & ALM
- Climate & Catastrophe Risk
- Model Validation
Sectors
Retail and commercial banks, life and P&C insurers, reinsurers, pension schemes, asset managers, and corporates with material treasury or insurance exposure.
Recent writing
All insights →AI Model Risk Management in 2026
AI adoption does not suspend model risk management. It raises the standard of evidence banks need before a model is allowed anywhere near a customer or a capital number.
PRA SS1/23 in Practice
SS1/23 is five principles long and deceptively demanding. The gap between a compliant framework and a working one is almost always ownership and evidence.
EU AI Act 2026: Transparency Duties
For financial institutions, the AI Act is less a new compliance regime than a documentation and classification problem attached to existing model governance.