You are an AI engineer
summary
Claim: web developers can become AI engineers without linear algebra or training models by calling LLM APIs and composing systems. Demonstrations use RAG (retrieval augmented generation) with a vector database (DataStax, Pinecone, Chroma, PGVector) plus GPT‑4 / embeddings-3 and LangFlow to ingest social posts and podcast transcripts, run semantic search, and return grounded answers. RAG addresses hallucinations, memory limits and knowledge cutoffs. Notes mention model sizes (7B, ~600B parameters), 2 million token context claims, and costs ~ $30–$60 per million tokens, with semantic caching to lower expenses.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 1:23Speaker introduction
- 2:27AI engineering defined
- 9:49AI beyond hype
- 15:02Generative AI applications
- 17:49Solving AI limits with RAG
- 27:00AI engineering challenges
- 30:28Summary and careers
related talks
38mYou Are an AI Engineer
37mBest AI Tools for Web Developers / How to Use AI in Web Development
41mAn Art, a Science, or Your Next Job Title?
48mGoing Beyond Prompt Engineering
25mBuilding Your own AI Assistant with JavaScript, LLMs, and Rag
41mFrom RAGs to riches: Build an AI document interrogation app in 30 mins
48m(Unpacking) AI's Impact on System Architecture
34mJavaScript-Powered AI: Crafting Intelligent Systems with LLMs and Vector Databases