Apppublic

Fatma033/technology_picker

sourceHugging Faceupdated 1y agoView on Hugging Face

0likes

App README

🚀 TechnologyPicker Crew

Welcome to the TechnologyPicker Crew project, powered by crewAI. This project sets up a multi-agent AI system designed to:

1.Find emerging technologies in a given sector using web search.
2.Research their adoption, benefits, risks, and growth outlook.
3.Select the most promising technology and generate a structured recommendation report.
4.Send the final decision report as a notification (email or push).

Our goal is to demonstrate how Agentic AI can accelerate innovation scouting, reduce manual research time, and improve decision-making for organizations.

⚙️ Installation

Ensure you have Python >=3.10 <3.13 installed. This project uses uv for dependency management.

Install uv if you don’t have it already:

bash

pip install uv

Then, from your project directory:

bash

uv sync

🔑 Configuration

Add your API keys in a .env file:

ini

OPENAI_API_KEY=your_openai_api_key
SERPER_API_KEY=your_serper_api_key
PUSHOVER_USER=your_pushover_user_key
PUSHOVER_TOKEN=your_pushover_token

▶️ Running the Project

You can run the crew in two ways:

1. From the terminal

bash

crewai run

2. With the Gradio Interface

We added a simple web interface for testing.

bash

uv run app.py

This will launch a Gradio app where you can:

—Enter a sector (e.g., Healthcare, IT, Energy).
—Run the workflow.
—View the structured report directly in your browser.
—Receive a final decision notification (email/push).

🧩 Crew Structure

The TechnologyPicker Crew is composed of specialized agents:

—Technology Trend Finder → searches the web for trending technologies.
—Technology Researcher → analyzes adoption, benefits, risks, and growth outlook.
—Technology Picker → selects the best technology and justifies the choice.
—Manager → coordinates the workflow, delivers the structured report, and triggers notifications.

Agents are defined in config/agents.yaml, tasks in config/tasks.yaml.

📂 Example Output

—output/trending_technologies.json → list of trending technologies.
—output/technology_research_report.json → structured research on each technology.
—output/decision.md → final recommendation (chosen technology + rationale).

📊 Benefits

—⏳ Time saved: research automated in minutes instead of hours.
—📑 Consistency: structured reports with clear adoption, risks, and outlook.
—📈 Scalability: adaptable to any sector (Healthcare, IT, Energy, Finance).
—📬 Actionable: results can be delivered as email or push notifications.