Big / Fast Data
From language models to intelligent agents
Language models such as GPT-3 and its successors continue to amaze us with their ever-expanding capabilities. However, what if we take things a step further by allowing these models to access data from various databases and granting them access to an array of tools such as web browsers, IDEs, and APIs? In this scenario, a language model could potentially become a genuine intelligent agent that can receive information from its environment, make decisions based on that information, and take actions that affect its surroundings.
During my talk, I will discuss a draft architecture for intelligent agents, as well as several recently developed tools that have made the development of such agents much more accessible. By leveraging these tools and resources, we can create intelligent agents that can assist us with complex tasks, streamline our workflows, and even automate entire processes.
- Speaker:
- Boris Cergol
- Podjetje
- Endava