Skip to main content
A specialized research crew built with CrewAI that leverages multiple agents to discover and analyze groundbreaking technologies. Demonstrates CrewAI’s multi-agent orchestration with sequential task processing.

​
Features

  • Multi-agent system architecture
  • Specialized researcher agents
  • Sequential task processing
  • Structured output generation
  • Powered by Meta-Llama-3.1-70B-Instruct

​
Prerequisites

​
Installation

1

Clone the repository

git clone https://github.com/Arindam200/awesome-ai-apps.git
cd starter_ai_agents/crewai_starter
2

Install dependencies

pip install -r requirements.txt
3

Configure environment

Create a .env file and add your Nebius API key:
NEBIUS_API_KEY=your_api_key_here

​
Implementation

​
Agent Configuration

Create a specialized researcher agent:
main.py
from crewai import Agent, Task, LLM, Crew, Process
import os
from dotenv import load_dotenv

load_dotenv()

# Create a researcher agent
researcher = Agent(
    role='Senior Researcher',
    goal='Discover groundbreaking technologies',
    verbose=True,
    llm=LLM(
        model="nebius/Qwen/Qwen3-235B-A22B",
        api_key=os.getenv("NEBIUS_API_KEY")
    ),
    backstory='A curious mind fascinated by cutting-edge innovation and the potential to change the world, you know everything about tech.'
)

​
Task Definition

Define research tasks with clear expectations:
main.py
# Task for the researcher
research_task = Task(
    description='Identify the next big trend in AI',
    expected_output='5 paragraphs on the next big AI trend',
    agent=researcher  # Assigning the task to the researcher
)

​
Crew Setup

Assemble the crew with sequential processing:
main.py
# Instantiate your crew
tech_crew = Crew(
    agents=[researcher],
    tasks=[research_task],
    process=Process.sequential  # Tasks will be executed one after the other
)

# Begin the task execution
tech_crew.kickoff()

​
Usage

Run the research crew: 
python main.py
The crew will execute the research task and provide detailed insights about emerging AI trends.

​
Example Research Topics

  • “Identify the next big trend in AI”
  • “Analyze emerging technologies in quantum computing”
  • “Research breakthroughs in sustainable tech”
  • “Investigate future of human-AI collaboration”
  • “Explore cutting-edge developments in robotics”

​
Technical Details

​
CrewAI Architecture

Agents

Specialized roles with defined goals and backstories

Tasks

Clear descriptions and expected outputs

Crew

Orchestrates agents and task execution

Process

Sequential or hierarchical execution

​
Agent Components

Agent(
    role='Senior Researcher',           # Agent's role
    goal='Discover groundbreaking technologies',  # Primary objective
    verbose=True,                       # Enable logging
    llm=LLM(...),                      # Language model
    backstory='...'                     # Context and personality
)

​
Task Components

Task(
    description='Identify the next big trend in AI',  # What to do
    expected_output='5 paragraphs...',               # Output format
    agent=researcher                                  # Who does it
)

​
Multi-Agent Patterns

​
Sequential Processing

Tasks execute one after another: 
crew = Crew(
    agents=[researcher, analyst, writer],
    tasks=[research_task, analysis_task, writing_task],
    process=Process.sequential
)

​
Hierarchical Processing

Manager agent coordinates worker agents: 
crew = Crew(
    agents=[manager, researcher, analyst],
    tasks=[task1, task2],
    process=Process.hierarchical,
    manager_llm=LLM(...)
)

​
Extending the Crew

​
Add More Agents

analyst = Agent(
    role='Data Analyst',
    goal='Analyze research findings and extract insights',
    verbose=True,
    llm=LLM(
        model="nebius/Qwen/Qwen3-235B-A22B",
        api_key=os.getenv("NEBIUS_API_KEY")
    ),
    backstory='Expert in data analysis with a keen eye for patterns and trends.'
)

writer = Agent(
    role='Technical Writer',
    goal='Create compelling reports from research data',
    verbose=True,
    llm=LLM(
        model="nebius/Qwen/Qwen3-235B-A22B",
        api_key=os.getenv("NEBIUS_API_KEY")
    ),
    backstory='Skilled writer who transforms complex technical information into clear, engaging content.'
)

​
Create Complex Workflows

research_task = Task(
    description='Research emerging AI trends',
    expected_output='Detailed research report',
    agent=researcher
)

analysis_task = Task(
    description='Analyze the research findings',
    expected_output='Data-driven insights',
    agent=analyst,
    context=[research_task]  # Depends on research_task
)

writing_task = Task(
    description='Write a comprehensive article',
    expected_output='Published article',
    agent=writer,
    context=[research_task, analysis_task]
)

crew = Crew(
    agents=[researcher, analyst, writer],
    tasks=[research_task, analysis_task, writing_task],
    process=Process.sequential
)

​
Add Custom Tools

from crewai_tools import tool

@tool("Search Tool")
def search_tool(query: str) -> str:
    """Search for information on the web"""
    # Implementation
    return results

researcher = Agent(
    role='Senior Researcher',
    goal='Discover groundbreaking technologies',
    tools=[search_tool],
    llm=LLM(...),
    backstory='...'
)

​
Best Practices

  • Give agents clear, specific roles
  • Define measurable goals
  • Write detailed backstories for context
  • Use verbose mode during development
  • Write clear task descriptions
  • Specify expected output format
  • Set realistic expectations
  • Use task context for dependencies
  • Choose appropriate process type
  • Consider task dependencies
  • Balance agent specialization
  • Test with different LLM settings

​
Workflow Visualization

​
Next Steps

Advanced Multi-Agent

Build complex multi-agent systems

RAG with CrewAI

Add knowledge base to your crew

Build docs developers (and LLMs) love

Get started for freeTalk to us