agent-opt
Python library. We’ll use the RandomSearchOptimizer
to keep things simple and demonstrate the core workflow.
1. Installation and Setup
First, install the library and set up your environment variables to connect to Future AGI for evaluations. You can get your API keys from the Future AGI dashboard.2. Prepare Your Dataset
Optimization is data-driven. You’ll need a dataset, which is a simple list of Python dictionaries. For this example, we’ll create a small dataset for a summarization task.3. Configure and Run the Optimization
Now, let’s set up the components and run the optimization. We’ll configure anEvaluator
to score our prompts, a DataMapper
to connect our data, and the RandomSearchOptimizer
to run the process.
4. Analyze the Results
Theresult
object contains the best prompt found and its final score.