LangGraph Agents and Workflows
This lesson demonstrates building and evaluating a LangChain agent for web research, specifically focusing on extracting structured information about individuals. The process leverages LangGraph for workflow management, LangSmith for evaluation and visualization, and LLMs for query generation and data analysis, showcasing a practical application of AI agents in data enrichment.
This lesson demonstrates building and deploying AI-powered applications using LangChain and LangGraph, showcasing the creation of a daily reflection bot that integrates with Slack. The instruction covers deployment processes, cron job scheduling, and leveraging LLMs for automated content generation and social media interaction.
This lesson demonstrates building a high-quality report-generation system using a multi-agent architecture powered by Llama 3.3 and NVIDIA NIM. The system prioritizes research before writing, leverages parallel processing for efficiency, and utilizes structured outputs for enhanced organization and report quality.
DeepSeek has released the open-source reasoning model, DeepSeek R1, trained using a novel reinforcement learning approach that prioritizes reasoning traces over simple next-token prediction. This innovative training, detailed in a companion paper, results in a model capable of complex reasoning tasks while offering various sizes for diverse computational resources.
This lesson explores building sophisticated AI agents using Large Language Models (LLMs), emphasizing simple, composable workflow patterns over complex frameworks. It covers various patterns like prompt chaining, parallelization, routing, and orchestration, showcasing how frameworks like LangGraph simplify development and deployment.
This lesson demonstrates building sophisticated multi-agent systems using Langchain's LangGraph, employing a supervisor agent to orchestrate specialized agents for complex tasks. The hierarchical structure, visualized through interactive diagrams and code examples, mirrors real-world organizational structures for efficient task delegation and problem-solving.
OpenAI's reasoning models, like the O3-mini, excel at complex tasks such as research and planning, unlike chat models; their efficient integration with other tools and cost-effectiveness make them valuable for diverse applications, pushing the boundaries of Artificial General Intelligence (AGI).
