{"id":36602,"date":"2025-04-01T19:09:31","date_gmt":"2025-04-01T13:39:31","guid":{"rendered":"https:\/\/mobisoftinfotech.com\/resources\/?p=36602"},"modified":"2025-10-15T16:38:18","modified_gmt":"2025-10-15T11:08:18","slug":"custom-ai-agent-development-crew-ai-framework","status":"publish","type":"post","link":"https:\/\/mobisoftinfotech.com\/resources\/blog\/ai-machine-learning\/custom-ai-agent-development-crew-ai-framework","title":{"rendered":"Custom AI Agent Development: How To Create AI Agents With Crew AI Agent Framework With Function Calling Support"},"content":{"rendered":"

In the last tutorial we saw <\/strong>how we can create an AI agent with Crew AI<\/strong><\/a> in pure python code. It’s advised for the reader to check that tutorial out to understand how Crew AI agents work at the code level.<\/p>\n\n\n\n

As your AI agents become more ambitious, you need proper structure to develop custom AI agents. The CrewAI framework offers a CLI utility that allows you to generate the scaffolding to jumpstart the creation of your AI agent. In this tutorial, we will create a simple Math Agent to demonstrate how to create AI agents using CrewAI CLI. Also, while developing custom AI agents, it’s a common requirement that you need to perform functions that the traditional LLM can\u2019t perform. To solve this, LLMs have introduced a concept of \u201cTool Calling<\/strong><\/a>\u201d or \u201cFunction Calling<\/strong><\/a>\u201d. Function Calling provides LLMs the power to perform tasks beyond just generating textual output. In this blog, we will also create a custom tool that the LLM can use to solve mathematical expressions. For businesses looking to streamline their AI integration, you may want to explore AI strategy consulting services<\/strong><\/a>, which can help align your AI efforts with long-term goals.  <\/p>\n\n\n\n

Note: All the code for this tutorial is present at this Github<\/a> repo.<\/strong><\/p>\n\n\n\n

Installation<\/strong><\/h2>\n\n\n\n

CrewAI requires Python >=3.10 and <3.13. Here\u2019s how to check your version:<\/p>\n\n\n

python3 --version<\/code><\/span>Code language:<\/span> Python<\/span> (<\/span>python<\/span>)<\/span><\/small><\/pre>\n\n\n
CrewAI<\/code> uses UV to manage dependencies. So we need to install UV. On Mac OS or Linux, you can use:<\/p>\n\n\n
curl -LsSf https:\/\/astral.sh\/uv\/install.sh | sh<\/code><\/span>Code language:<\/span> Python<\/span> (<\/span>python<\/span>)<\/span><\/small><\/pre>\n\n\n
Once the uv command is installed, you can use the following command to install the crewai<\/code> CLI.<\/p>\n\n\n
uv tool install crewai<\/code><\/span>Code language:<\/span> Python<\/span> (<\/span>python<\/span>)<\/span><\/small><\/pre>\n\n\n
Now, we can create our project<\/p>\n\n\n
crewai create crew math-agent<\/code><\/span>Code language:<\/span> Python<\/span> (<\/span>python<\/span>)<\/span><\/small><\/pre>\n\n\n
Select ‘OpenAI’ as the provider and choose ‘gpt-4o’ as the model. Since we are going to use function calling in this tutorial ,it’s important to choose a model that has function calling capabilities.<\/p>\n\n\n\n
This should create a directory structure like the following<\/h3>\n\n\n
math_agent\n\u251c\u2500\u2500 README.md\n\u251c\u2500\u2500 knowledge\n\u2502   \u2514\u2500\u2500 user_preference.txt\n\u251c\u2500\u2500 pyproject.toml\n\u251c\u2500\u2500 src\n\u2502   \u2514\u2500\u2500 math_agent\n\u2502       \u251c\u2500\u2500 __init__.py\n\u2502       \u251c\u2500\u2500 config\n\u2502       \u2502   \u251c\u2500\u2500 agents.yaml\n\u2502       \u2502   \u2514\u2500\u2500 tasks.yaml\n\u2502       \u251c\u2500\u2500 crew.py\n\u2502       \u251c\u2500\u2500 main.py\n\u2502       \u2514\u2500\u2500 tools\n\u2502           \u251c\u2500\u2500 __init__.py\n\u2502           \u2514\u2500\u2500 custom_tool.py\n\u2514\u2500\u2500 tests\n\n7<\/span> directories, 10<\/span> files\n<\/code><\/span>Code language:<\/span> Python<\/span> (<\/span>python<\/span>)<\/span><\/small><\/pre>\n\n\n
pyproject.toml<\/strong>
A standardized configuration file that holds project metadata, dependencies, and build configuration. It\u2019s used by modern Python packaging tools (like uv) to manage your project environment.<\/p>\n\n\n\n
Directory src\/math_agent\/<\/code><\/strong><\/h3>\n\n\n\n
config\/<\/strong><\/h4>\n\n\n\n
Contains YAML configuration files that define how your AI agents and their tasks behave:<\/p>\n\n\n\n