AI in Practice - Building Real-World AI Systems with Modern Frameworks

Learn to build practical AI systems through a comprehensive video series covering advanced frameworks and implementation techniques. Explore NVIDIA NeMo Guardrails for creating secure chatbots with variables, flows, and action-based guidance systems. Master fine-tuning techniques for OpenAI's GPT 3.5 within LangChain agent architectures and discover how to optimize RAG (Retrieval-Augmented Generation) chatbots for speed and performance. Dive into alternative AI solutions including Cohere Embed v3 and open-source options, while implementing the Semantic Router framework for steerable chatbot development. Gain hands-on experience with faster LLM function calling through dynamic routes and local implementation using Llama.cpp. Compare OpenAI's latest embedding models, including the new 256-dimensional embeddings versus Ada 002, and learn advanced evaluation techniques using RAGAS for AI agents. Master semantic chunking methods to improve RAG performance and explore NVIDIA's AI Workbench for streamlined AI engineering workflows. Process videos for GPT-4o integration and search capabilities while implementing three distinct semantic chunking approaches. Build sophisticated agents using LangGraph's deep architecture and create asynchronous AI agents with Llama Index Workflows. Conclude with advanced guardrail implementation for AI agent safety and reliability in production environments.