Agents by Chip Huyen

In traditional software engineering, conditions for control flows are exact. With AI-powered agents, AI models determine control flows. Plans with non-sequential control flows are more difficult to both generate and translate into executable commands. (View Highlight)

First proposed by ReAct (Yao et al., 2022), interleaving reasoning and action has become a common pattern for agents. Yao et al. used the term “reasoning” to encompass both planning and reflection. At each step, the agent is asked to explain its thinking (planning), take actions, then analyze observations (reflection), until the task is considered finished by the agent. The agent is typically prompted, using examples, to generate outputs in the following format: (View Highlight)

Compared to plan generation, reflection is relatively easy to implement and can bring surprisingly good performance improvement. The downside of this approach is latency and cost. Thoughts, observations, and sometimes actions can take a lot of tokens to generate, which increases cost and user-perceived latency, especially for tasks with many intermediate steps. To nudge their agents to follow the format, both ReAct and Reflexion authors used plenty of examples in their prompts. This increases the cost of computing input tokens and reduces the context space available for other information. (View Highlight)

More tools give the agent more capabilities. However, the more tools there are, the harder it is to efficiently use them. It’s similar to how it’s harder for humans to master a large set of tools. Adding tools also means increasing tool descriptions, which might not fit into a model’s context. (View Highlight)

Vogager (Wang et al., 2023) proposes a skill manager to keep track of new skills (tools) that an agent acquires for later reuse. Each skill is a coding program. When the skill manager determines a newly created skill is to be useful (e.g., because it’s successfully helped an agent accomplish a task), it adds this skill to the skill library (conceptually similar to the tool inventory). This skill can be retrieved later to use for other tasks. (View Highlight)

agents also have unique failures caused by planning, tool execution, and efficiency. Some of the failures are easier to catch than others. (View Highlight)

At its core, the concept of an agent is fairly simple. An agent is defined by the environment it operates in and the set of tools it has access to. In an AI-powered agent, the AI model is the brain that leverages its tools and feedback from the environment to plan how best to accomplish a task. (View Highlight)