Roadmap
Last updated
Last updated
Below is our planned major feature roadmap for the Great Wave AI Platform. Minor features will be introduced alongside these updates to continuously enhance the platform's functionality and user experience. This roadmap is designed to evolve based on user needs and feedback, so if you are reading this and have ideas or suggestions, please don’t hesitate to get in touch - . Your input is invaluable in shaping the future of the platform!
Q1 2025
Q2 2025
Adaptive Agent Beta
The Adaptive Agent is designed to dynamically coordinate multiple agents based on the input it receives. By evaluating the context and requirements of the input, it autonomously selects and invokes the most relevant agents, enabling efficient and context-aware handling of complex tasks.
Agent Memory
Agents will feature configurable memory, enabling them to retain context and information within conversational interfaces. This allows for more dynamic and natural interactions, as agents can reference prior exchanges and adapt responses based on the conversation history.
Agent Validation
Agents will include self-validation capabilities, ensuring their outputs meet predefined criteria before proceeding. This feature is especially valuable in multi-agent architectures, where it helps maintain integrity and accuracy. If validation criteria are not met, the process chain will pause, enabling troubleshooting or refinement before continuing.
Evaluation Improvements
The Evaluation module will include visibility into Triggered Guardrails and Agent Validation processes, enhancing transparency and debugging in multi-agent architectures. Additionally, UI enhancements provide a seamless way to view and navigate chunks across complex workflows, making evaluation and monitoring more intuitive and efficient.
Instructions Versioning
Agent Instructions will support version control, enabling users to track changes and revert to previous versions as needed. This feature ensures flexibility and reliability, making it easier to experiment, refine, and maintain consistency in agent behavior over time.
Instructions UI Changes
The Instructions Module will be updated with improvements to deliver a more seamless and intuitive user experience. These changes simplify navigation, enhance clarity, and streamline the process of creating, editing, and managing agent instructions.
Document Modification Handling
Knowledge Collections will support overwrites, and auto multi index refresh for both through the API and within the UI/UX. This enhancement ensures flexibility in updating documents, allowing users to efficiently modify content while maintaining consistency across the system.
Observe Domain
A centralized dashboard providing a comprehensive view of all agent activity in a single interface. This new area enables users to monitor, analyze, and manage agent operations seamlessly, offering real-time insights and improving overall visibility into system performance.
Publishing
Each organization will have access to dedicated development and live environments, ensuring a seamless workflow for building, testing, and deploying agents. Teams can easily create, refine, and validate agents in the development environment before publishing them to the live environment with confidence. This structured approach promotes collaboration, reduces errors, and streamlines the transition from prototype to production.
Neuro-Symbolic AI
Enable users to create and interact with property graphs, a powerful tool for representing and analyzing complex relationships. These graphs consist of labeled nodes (representing entities, categories, or text labels) enriched with properties (metadata) and connected through relationships that form structured paths. By leveraging a Property Graph Index, users can seamlessly organize, visualize, and query interconnected data, uncovering patterns and insights with the power of neurosymbolic AI. This integration combines symbolic reasoning with neural adaptability, making it ideal for exploring dynamic and intricate knowledge collections.
Consistency Evaluation
Introducing tools to assess and ensure agent consistency. Users can select specific interactions to test for consistency and access global consistency metrics for each agent. These features help identify discrepancies, improve reliability, and maintain a cohesive agent experience.