What an Agent Harness Needs Before Production: Eleven Failures and the Part That Stops Each

An agent harness is the code around the model. A bare twenty-line loop meets eleven bad days: a runaway model, bad arguments, a slow tool, a hidden instruction, generated code that reads secrets, a crash after a refund. Then the eight parts that stop them, built with JSON Schema, SQLite, OpenTelemetry, process limits and a container.

AI/ML14 min read

Taking a GenAI Assistant to Production on Amazon Bedrock: Knowledge Base, Fine-Tuning and Release Gates for Irish Healthcare

For an Irish health guidance assistant on Amazon Bedrock, published HIQA and HSE guidance goes into a Knowledge Base in eu-west-1, fine-tuning is kept for behaviour on non-personal data because Nova customisation runs in US Regions, and IHI and PPSN values are blocked by a dataset gate and a Guardrail. A Step Functions release pipeline promotes a change only after a golden-set evaluation and clinical sign-off.

AI/ML18 min read

Testing with LocalStack: AWS on your Laptop

If you build on AWS, you need **LocalStack**. It mocks almost every AWS service (S3, Lambda, DynamoDB, SQS) on your local machine via Docker, allowing for free, fast, and offline development. Docker…

AWS1 min read

Integration Testing with Testcontainers

Mocking the database in tests is a lie. In-memory databases (like EF Core InMemory) behave differently than SQL Server. **Testcontainers** allows you to spin up real, throwaway Docker containers for your integration…

PEX – Automated White Box Testing for .NET

  Automated White Box Testing for .NET Pex (Program EXploration) produces a traditional unit test suite with high code coverage. A parameterized unit test is simply a method that takes parameters, calls…

.NET1 min read