Course 4Production Actuarial Systems, Software Engineering & Enterprise Deployment
Master software design patterns, automated testing, version-controlled CI/CD pipelines, modern web applications/dashboards, and actuarial simulation engines (e.g., life contingencies, economic capital modeling).
For: Students who have mastered actuarial data science, machine learning, and modeling, and want to learn how to turn notebooks into robust, tested, and deployable software applications.
-
Lesson 1
Software Architecture: from Script to Package
src/ layout, pyproject.toml, type hints, pydantic validation, Strategy and Factory patterns. Lab: refactor a monolithic pricing script.
-
Lesson 2
Automated Testing, Code Quality & CI
pytest, TDD, hypothesis, coverage and mutants, ruff, mypy, GitHub Actions. Lab: a test suite for a multi-coverage pricing engine.
-
Lesson 3
Stochastic Simulation & Performance
Seeds, profiling, vectorization, numba. Lab: a million-scenario annuity ALM model and its economic capital.
-
Lesson 4
Actuarial APIs with FastAPI
Schemas, validation, status codes, API keys, OpenAPI, async. Lab: a live quote service for the Course 3 champion model.
-
Lesson 5
Interactive Dashboards with Streamlit
Reruns, caching, session state, uploads, AppTest. Lab: the Executive Underwriting Portal.
-
Lesson 6
Enterprise Data Stores: SQL & SQLAlchemy
Keys and constraints, transactions, injection, chunking, the ORM, SQLite and PostgreSQL. Lab: CSV pipeline to SQL.
-
Lesson 7
Containerization with Docker
Multi-stage images, configuration and secrets, Docker Compose. Lab: the API and portal in containers.
-
Lesson 8 · Part A
Integration, Delivery & Monitoring
Verified model artifacts, promotion rules, a CI/CD pipeline with smoke tests, drift monitoring.
-
Lesson 8 · Part B
Capstone: An Enterprise Actuarial Rating Platform
The champion model behind an API, a database, an underwriting portal, containers, and CI/CD.