Customizing Foundation Models

The GenAI heart of MLA-C02: prompt engineering as a customization layer, fine-tuning without catastrophic forgetting, continued pre-training, distillation, and retrieval optimization.

Content current as of 2026-10.

Lessons

1. Prompt engineering as a customization layer
2. Fine-tuning strategies and catastrophic forgetting
3. Continued pre-training and distillation
4. Optimizing retrieval: embeddings and rerankers

Sources

• Amazon Bedrock — custom models (fine-tuning and continued pre-training)
• Amazon Bedrock — prompt engineering guidelines
• Amazon Bedrock — model distillation
• Amazon Bedrock Knowledge Bases