AI A-Z
summary
Led an EPAM engineering team building dozens of AI prototypes and explains why size matters citing 178 billion parameters in GPT‑3, shows concrete tooling: vector databases, retrieval‑augmented generation (RAG) pipelines, LongChain/agent workflows and AutoGen, and compares Java libraries (DL4J, TensorFlow Java) and low‑level work (Panama, vector units). Gives measured costs and limits — example: a generated financial report can cost $5 per run — and calls out instability, nondeterminism and performance tradeoffs so you know practical risks and wins.
AI-written summary · about
chapters (22)
- 0:00Introduction
- 1:10AI A-Z overview
- 2:34What is AI
- 4:23AI history
- 7:18Computational power
- 9:27Neural networks
- 14:14Why size matters
- 16:01Deep learning & LLMs
- 20:27Transformers & attention
- 22:16Early AI models
- 25:08Modern AI & generative hype
- 26:16Retrieval augmented generation explained
- 29:26AI agents & LangChain
- 33:02Java to AI paths
- 35:29AI developer roles
- 37:13AI hardware & CPU
- 39:21AI apps & prompt engineering
- 41:05AI hype & economics
- 46:24AI challenges today
- 49:15Right tool for AI
- 50:38Recap
- 51:20Q&A
related talks
47mThe Rise of AI: A Reality Check on Generative AI
19mLet's Talk GenAI
49mGenerative AI: Concepts, Tools, and Applications for Data Professionals
25mThe Great AI Game
37mGenerative AI: Beyond (Just) Hype
19mNow in AI - How We Got Here
45mBringing the Power of AI to Your Application
15mHarnessing the Power of Generative AI for Intelligent Applications