Tech Stack
- Agno framework for AI agent development
- Nebius AI’s for running LLMs
- Duckduckgo for web search
- Yahoo Finance for finance data
- Couchbase as vector storage
Prerequisites
- Python 3.10 or higher
- Nebius API key (get it from Nebius AI Studio)
- Couchbase database credentials (for couchbase example)
Setup
-
Get the code:
-
Install dependencies:
usinguv
:or install using python pip: -
Create .env file in the project root and add your Nebius API key:
Running the Agent
Usinguv
:
- select the kernel we defined earlier
- run the notebook
Agents
Web search agent
Uses duckduckgo to search the web agno_websearch_agent.ipynb- Creating an agent with a knowledge base for specialized domains
- Building multi-agent systems where specialized agents work together
Multi Agent System
This example shows how mutiple agents work together.- A web search agent for finding general information
- A finance agent for retrieving financial data
- A coordinator agent that delegates tasks to the specialized agents
Knowledgebase Agent
This example demonstrates how to create an agent with specialized knowledge. Use couchbase as vector database. Prerequisites- We need to connect to a couchbase instance. You can sign up for a free account at www.couchbase.com
-
Also add the following attributes to
.env
file:
References and Acknowledgements
- Agno Framework
- Nebius AI
- Contributed from awesome-ai-apps