Mydra logo
Log in
>All courses>Artificial Intelligence
Artificial Intelligence
DeepLearning.AI logo

DeepLearning.AI

Function-Calling and Data Extraction with LLMs

  • up to 1 hour
  • Intermediate

This course teaches you to extend LLMs with custom functionality via function-calling and extract structured data from natural language inputs. Build an end-to-end application that processes customer service transcripts using LLMs.

  • Function-calling
  • Structured data extraction
  • NexusRavenV2-13B
  • OpenAPI specifications
  • SQL

Overview

In this course, you will learn two critical skills for building applications with LLMs: function-calling and structured data extraction. You will work with NexusRavenV2-13B, an open-source model fine-tuned for these tasks. The course covers forming prompts with function definitions, using LLM responses to call functions, handling multiple function calls, and building applications that process customer service transcripts. These skills will enable you to create advanced AI agents and assistants for various real-world applications.

  • Web Streamline Icon: https://streamlinehq.com
    Online
    course location
  • Layers 1 Streamline Icon: https://streamlinehq.com
    English
    course language
  • Self-paced
    course format
  • Live classes
    delivered online

Who is this course for?

AI Enthusiasts

Individuals interested in using function-calling capabilities with large language models to interface with external tools and extract structured data.

Developers

Developers looking to build advanced AI agents and assistants that can process and analyze customer feedback, automate data entry, and enhance search and recommendation systems.

Data Scientists

Data scientists aiming to extend LLMs with custom functionality via function-calling and extract structured data from natural language inputs.

This course will teach you to extend LLMs with custom functionality and extract structured data from natural language inputs. Ideal for AI enthusiasts, developers, and data scientists, it will help you build advanced AI agents and assistants for various real-world applications.

Pre-Requisites

1 / 2

  • Familiarity with LLMs

  • Basic Python knowledge

What will you learn?

Introduction to Function-Calling
Learn how to extend LLMs with custom capabilities by enabling them to form calls to external functions based on natural language instructions.
Structured Data Extraction
Understand how to pull usable information from unstructured text using LLMs.
Working with NexusRavenV2-13B
Explore the NexusRavenV2-13B model, fine-tuned for function-calling and data extraction, and learn how to use it effectively.
Forming Prompts with Function Definitions
Learn to form prompts with function definitions and use LLM responses to call those functions.
Handling Multiple Function Calls
Use an LLM with multiple function calls, including parallel and nested function calls, to create complex agent workflows.
Using OpenAPI Specifications
Build function calls that can access web services using OpenAPI specifications.
Building an End-to-End Application
Build an application that processes customer service transcripts, generates SQL calls, and stores results in a database with commands generated by the LLM.

Meet your instructors

  • Jiantao Jiao

    Assistant Professor, Nexusflow AI

    Jiantao Jiao conducts research in large language models at UC Berkeley and collaborates with researchers and engineers to develop and democratize generative AI through Nexusflow.

  • Venkat Srinivasan

    Founding Engineer (NLP), Nexusflow

    Venkat Srinivasan is a Founding Engineer at Nexusflow, specializing in natural language processing (NLP). He brings diverse experience from previous roles at Huawei, Oracle, and NVIDIA, where he worked on projects related to CPU architecture testing, simulation, and software development.

Upcoming cohorts

  • Dates

    start now

Free