Changelog
Discover what's new!
Stay up to date with the latest enhancements and updates
EdFlow: Enhancing Agent Training with Curated Datasets
High-quality training data is essential for developing effective AI agents. EdFlow now provides Datasets, allowing you to curate collections of data specifically for agent training. By tagging examples as "good" or "bad," you help the agent learn from both positive and negative instances.
For instance, when reviewing agent outputs, you might find responses that meet your criteria. Tagging these as "good" includes them in the training data as positive examples. If an output doesn't meet your standards, tagging it as "bad" adds it as a negative example, teaching the agent to avoid similar mistakes in the future.
By building datasets with both positive and negative examples, you can improve your agents' performance based on actual interactions and your specific guidelines.
EdFlow: Monitoring and Refining Agents with Runs
Understanding how your AI agents perform is important for delivering high-quality services. EdFlow introduces Runs, which record your agents' activities by capturing inputs, outputs, and other relevant data.
You can review these Runs
to see how your agents handle tasks. If you notice an agent's response that needs improvement, you can add the input and output from that run directly to your Datasets. By tagging these examples appropriately, you contribute to the agent's training data, refining its performance based on actual usage.
By using Runs
and integrating them into your Datasets
, you can continuously enhance your agents to better serve your educational objectives.