Asana's AI agents share memory across your company — but not your secrets

Asana's Agentic Work Management (AWM) operating system enables AI agents to share memory across a company while maintaining data governance and access controls. AWM is built on top of Asana's 18-year-old Work Graph, a graph-based database that organizes information through a Pyramid of Clarity structure. This allows AI agents to view overarching company goals, update project statuses, and share memory with human colleagues. The system has been deployed to several customers, including FedEx, and has solved technical hurdles such as data governance and dynamic model routing. The practical implication for engineers building AI systems is the ability to create multiplayer teammates that can operate alongside humans, sharing memory and learning from workflows and human feedback.

⚡ Key Takeaways

• Asana's AWM operating system is built on top of their 18-year-old Work Graph, a graph-based database that organizes information through a Pyramid of Clarity structure.
• The Work Graph provides a real-time ledger of who does what, by when, and why, and allows AI agents to view overarching company goals, update project statuses, and share memory with human colleagues.
• AWM handles dynamic model routing to abstract prompt engineering away from the user, and has a system of access controls to govern what triggers the creation of a memory versus the simple execution of a task.
• The system has been deployed to several customers, including FedEx, and has solved technical hurdles such as data governance and dynamic model routing.
• AWM requires a ready-made enterprise context graph, such as the Work Graph, to operate effectively.

The development of AWM has significant implications for engineers building AI systems, as it enables the creation of multiplayer teammates that can operate alongside humans, sharing memory and learning from workflows and human feedback. This can lead to more efficient and effective collaboration between humans and AI agents, and can help to solve real-world bottlenecks in enterprise environments.

✅ Practical Steps

1. Apply the concepts from this article to your own system design, considering the use of graph-based databases and access controls to govern AI agent memory and behavior.
2. Explore the use of dynamic model routing to abstract prompt engineering away from the user, and consider how this can be applied to your own AI systems.
3. Consider the importance of data governance and access controls in AI system design, and how these can be used to ensure the secure and effective operation of AI agents in enterprise environments.

Want the full story? Read the original article.