# Day 1 Lecture: Trusting your workflows (evaluation metrics and guardrails)

> Understanding evaluation metrics and guardrails is crucial for ensuring the quality of machine learning applications in production. This lecture covers the common pitfalls of deploying large language models, including issues like hallucination and prompt injection, and discusses strategies to preve…

- Web page: https://www.dataexpert.io/lesson/ml-ops-day1-lecture-july25-p737-l1183
- Program: [AIExpert](https://www.dataexpert.io/program/ai-expert)
- Module: Week 4: MLOps, Guardrails & Business Value
- Access: Requires enrollment in AIExpert
- Length: 27 min video
- Skills: NLP, MLOps, LLMs
- Academy: DataExpert.io Academy

## About this lesson

Understanding evaluation metrics and guardrails is crucial for ensuring the quality of machine learning applications in production. This lecture covers the common pitfalls of deploying large language models, including issues like hallucination and prompt injection, and discusses strategies to prevent these problems using techniques such as multi-tenant architecture and input/output guardrails. Data scientists and AI engineers will benefit from real-world examples and practical approaches to safeguard their applications.

## Navigation

- Previous lesson: [MLOPs and Guardrails Day 2 Lab](https://www.dataexpert.io/lesson/ml-ops-day2-lab-oct25-p737-l1525.md)
- Next lesson: [Day 1 Lab: Setting up CI/CD for an AI application](https://www.dataexpert.io/lesson/ml-ops-day1-lab-july25-p737-l1184.md)