function calling<\/a> to interact with systems (such as APIs or databases) often fall short of this goal. They lack the robustness, controllability, and built-in support for complex processes required by enterprise-grade applications.<\/p>\nIn this article, we\u2019ll explore why this is the case and introduce an alternative approach: process calling. <\/strong>This enables the creation of reliable, process-aware, and easily debuggable conversational agents. We\u2019ll also share code examples and walk you through how to get started with the Rasa platform.<\/p>\nIn the current paradigm, AI agents are equipped with tools that enable them to solve specific tasks. These tools typically perform atomic actions, such as calling an API to read or write data, updating or fetching data from a database, or similar operations. The limitation of such an approach is that it often lacks state<\/em>, making AI agents unpredictable and sometimes even unreliable for several reasons:<\/p>\n\n- Loss of conversational context:<\/strong> The agent doesn\u2019t remember previous conversations or decisions, leading to redundant or inconsistent responses.<\/li>\n
- Poor adherence to business processes:<\/strong> Without state tracking, the agent may skip required steps or follow steps in the wrong order.<\/li>\n
- Inconsistent execution of repeated tasks:<\/strong> The same task may yield different results, breaking user expectations and reducing trust.<\/li>\n<\/ul>\n
On the other hand, businesses have well-established processes, and AI assistants are expected to follow them, not improvise or create their own. A conversational AI agent deployed for customer service must understand users\u2019 needs, connect them to the right company processes, clearly explain how it can assist, and guide them through each step to achieve their goals\u2013all while maintaining a smooth and natural conversation flow.\u00a0<\/p>\n
This is where process calling<\/strong> comes in. With process calling, the LLM invokes and collaborates with a stateful process. The user asks the assistant a question, and the LLM predicts which specific, defined business process to trigger. The process, along with LLM, works together to drive the conversation forward.\u00a0<\/p>\nLet\u2019s dive into how to build a reliable AI assistant using the process-calling approach in practice. We will develop a banking AI agent capable of handling simple processes, including transferring money, opening a savings account, responding to frequently asked questions (FAQs), and addressing off-topic requests.<\/p>\n
\n