AI observability refers to the ability to gain insights and understand the behavior, performance, and cost of artificial intelligence (AI) models and services during their operation. This involves monitoring, analyzing, and visualizing the important internal states, inputs, and outputs of AI models to ensure their correctness, reliability, and effectiveness.
In this Observability Clinic with Andreas Grabner @andreas_grabner, we have Wolfgang Beer @wolfgang_beer , Principal Product Manager, walk us through how to use Dynatrace to monitor the usage of AI APIs (such as OpenAI, TensorFlow, or others), identify costs and how to diagnose and optimize performance, and costs.
In this session, we discussed the following links:
AI Observability Blog: https://dynatr.ac/3PabDER
AI Observability Doc: https://dynatr.ac/463qbfx
GitHub Repository for OpenAI: LINK
GitHub Repository for TensorFlow: LINK
BizEvents Doc: https://dynatr.ac/468CsiA
Chapter List:
00:00 - Introduction
01:02 - What you'll learn today
02:47 - OpenAI Demo Environment
05:52 - How to capture AI APIs context
09:47 - Log Parsing Rules
11:40 - Live Demo - OpenAI Observability
25:48 - TensorFlow Observability
28:06 - Live Demo - TensorFlow Observability
- - -
Subscribe to our YT channel
Stay up-to-date with Dynatrace! Follow us on Facebook, Instagram, LinkedIn, Twitter, Twitch
⭐ When passion meets people magic and innovation happen. ⭐
