AI Model Risk – Associate
A role in the firm's independent model risk group, validating the AI and machine-learning models used across the bank. You work alongside AI developers and researchers to challenge their methods.
What you’d work on
Our reading of the posting: what a new hire in this role would most likely spend their first months on.
- Independently test accuracy, explainability and robustness of AI models
- Build benchmark models to compare against the model under review
- Write up limitations and risks found during testing
- Discuss findings with model developers and agree remediation
What they ask for
Required
- 2-5 years in AI or ML validation or research
- Python with NumPy and Pandas
- TensorFlow or PyTorch
- Statistical modelling
- Machine learning algorithms
- Strong communication
- Education
- Bachelor's, Master's or PhD in computer science, maths, physics, engineering or a similar field
- Experience
- 2-5 years in AI/ML model validation or research
- Languages and tools
How to prepare
- Train a small model and stress-test it for drift, leakage and unstable explanations
- Learn explainability methods such as SHAP and partial dependence, and where they mislead
- Build a simple benchmark model and compare it against a more complex one
- Practise writing a short validation report that states limitations plainly
What to study
- Financial Machine Learning
- Quantitative Risk Management
- Probability, Statistics and Financial Data
- Python for Quantitative Finance
- Professional Quant Practice
The Model Validation Quant roadmap sets out the stages for this kind of role, with free resources for each.
Open the roadmapApply
Read the full posting on Goldman Sachs’s site before applying: requirements, location and pay can change after we check them.
View the posting on Goldman Sachs’s site ↗
SelfMadeQuant wrote this summary from the employer’s public posting, last checked 6 November 2026. SelfMadeQuant isn’t affiliated with Goldman Sachs and doesn’t handle applications.