Completed
07:21 Technology issues
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
GenAI Grows Up - Building Production-Ready Agents on the JVM
Automatically move to the next video in the Classroom when playback concludes
- 1 00:00 Intro
- 2 01:53 Let's make this personal
- 3 02:24 Tools add power
- 4 03:36 "Personal assistant" use case
- 5 04:30 Powerful but not predictable is sometimes OK
- 6 06:45 Alarming failure rate
- 7 07:19 Why do GenAI projects fail?
- 8 07:21 Technology issues
- 9 09:56 Prompt engineering: The new alchemy
- 10 12:48 Integration problems
- 11 13:28 Excessive vendor influence
- 12 13:57 Top-down mandates
- 13 15:15 Organizational issues
- 14 15:55 How do we fix this?
- 15 16:04 1. Attack nondeterminism
- 16 21:11 2. Integrate with what works
- 17 22:30 Bringing structure to LLM interactions
- 18 23:42 Domain-integrated context engineering
- 19 26:47 What is the role of Java developers?
- 20 27:27 Option 1: Imitate Python approaches
- 21 28:20 Option 2: Build something better
- 22 29:10 The stakes are high
- 23 29:46 What the JVM brings to GenAI
- 24 30:32 Python vs JVM for enterprise AI
- 25 32:22 Time for Java community to lead again
- 26 33:01 Introducing Embabel
- 27 39:19 Java example: Bank support agent
- 28 41:40 Seamless code actions & LLM invocation
- 29 42:29 Builds on Spring
- 30 44:03 Modern Java/Kotlin API
- 31 46:01 Example: This slide deck
- 32 49:24 Decker agent example
- 33 50:04 Embabel vs Python frameworks
- 34 51:22 Conclusion
- 35 52:38 Start right away
- 36 53:32 Outro
