World's Fair 2024
Build an AI Research Agent: Apoorva Joshi
Overview
This talk introduces the fundamental concepts of AI agents, detailing their components and use cases. It emphasizes that agents are best suited for complex, multi-step tasks requiring the integration of various capabilities, such as data aggregation, visualization, and reasoning, or when personalization and adaptive responses are necessary. The session also provides a hands-on guide to building a research agent from scratch, incorporating tools, memory, and reasoning patterns.
Who should watch
- AI Engineers
- Product Managers
- Builders working with LLMs
- Those looking to create agents for complex tasks
- Developers interested in research assistance tools
Key takeaways
- AI agents use LLMs to reason, plan, and execute tasks with the help of tools and memory.
- Agents differ from simple prompting and RAG by enabling tool use, iterative reasoning, and personalization.
- Key components of agents include planning and reasoning (e.g., Chain of Thought, Tree of Thoughts, ReAct, Reflection), memory (short-term and long-term), and tools for external interaction.
- Agents are ideal for complex, multi-step tasks that require integrating multiple capabilities or personalization.
- The workshop demonstrates building a research agent using LangChain, integrating tools for paper summarization, paper retrieval, and question answering with a MongoDB knowledge base.
- The agent is enhanced with short-term memory using chat message history persisted in MongoDB.
- LangChain Expression Language (LCEL) is used for declarative chaining of prompts, LLMs, and tools.
Notable quotes
*An AI agent is a system that uses a large language model or LLM to reason through a problem create a plan to solve the problem and also execute the plan with the help of a set of tools.*
*The tldr is use agents for complex multi-step tasks that require integration of multiple capabilities such as question answering task execution analysis that kind of thing and using all of these to arrive at a final answer or outcome and also if there is a need for personalization or adapted responses.*
Unofficial community note. Prefer the recording for nuance.