Streamlined Implementation: AnswerRocket's Latest Update
We’re excited to announce a new update designed to make implementing your instance of Max smoother and more efficient. This release focuses on providing analysts with tools to simplify configuration, enhance usability, and improve monitoring capabilities. Here’s what’s new:
Key Updates:
1. Validate LLM Keys
- Per-field Validation Rules: New validation rules have been added to the API configuration forms to catch errors such as out-of-range values or missing required fields.
- Validate Button: A "Validate" button has been added to allow users to check their inputs with a real LLM request before saving changes. This helps ensure accuracy and save time during setup.
2. Scheduler UI Improvements
- JSON Code Editor: Schedules can now be configured directly using a JSON code editor, removing the need to upload files. This change provides a more intuitive and flexible way to manage schedules. In addition this makes it easy to know and modify the current configuration.
3. Document Methods for Skill Platform
- Documentation Access: In Skill Studio, a new question mark button opens a panel displaying the skill platform documentation. Users can also choose to open the documentation in a new window for easier reference.
4. Monitoring Max
- Enhanced Data: Monitoring your Max copilot engagement is now more detailed, with the addition of concatenated full names and the time a question takes in seconds. This helps in better tracking and understanding user interactions.
5. Caching Options
- Copilot-Level Caching: There is now an option to enable or disable caching at the copilot level. This is particularly useful for real-time data use cases, allowing for more responsive and dynamic interactions.
6. Copilot Question Settings
- Tabbed Management: Properties for questions can now be managed within tabbed sections, streamlining the user interface and improving workflow efficiency.
7. Auditing Enhancements
- Dataset Reload Auditing: When a dataset is reloaded, this action is now captured in the audit trail. This feature enhances traceability and accountability within your data management processes.
These updates are designed to provide a smoother, more intuitive experience for analysts implementing Max. We hope these enhancements will save you time and reduce the complexity of managing your instance.
In addition we resolved an issue where any explanation provided by the language model at the time of running a skill would previously disappear after first being rendered. That text is now persisted.
Tested Models
Chat: gpt-4o-2024-05-13
Narrative: gpt-4o-2024-05-13
Embedding: text-embedding-3-small
Updated