The previous thread glossed over how our LLM Agents actually work.
The truth is, it took us a long time to figure out how to get reliable and impressive results from agents.
By the end, we learned general strategies to build effective LLM agents, which we're now sharing. 🧵
The truth is, it took us a long time to figure out how to get reliable and impressive results from agents.
By the end, we learned general strategies to build effective LLM agents, which we're now sharing. 🧵
https://twitter.com/tjbecker_/status/1955678204498440292
Strategy #1: Decompose the task
LLM agents excel at tasks generally requiring human intuition to solve, but they can't yet solve arbitrarily complex multi-step tasks. If the task can be solved in multiple parts, you should decompose it as a workflow of multiple agents.
LLM agents excel at tasks generally requiring human intuition to solve, but they can't yet solve arbitrarily complex multi-step tasks. If the task can be solved in multiple parts, you should decompose it as a workflow of multiple agents.
Strategy #2: Curate the Toolset
LLM agents repeatedly call tools until they reach their goal, so curating the toolset is crucial.
The toolset should be as powerful, focused, and helpful as possible. Put up guardrails to prevent your agents from reaching known dead-ends!
LLM agents repeatedly call tools until they reach their goal, so curating the toolset is crucial.
The toolset should be as powerful, focused, and helpful as possible. Put up guardrails to prevent your agents from reaching known dead-ends!
Strategy #3: Structure Complex Outputs
Make sure your agent knows exactly what it needs to output, including the precise format of that output. Pro tip: you can ask them to output information which you don't plan to use, but that steers them towards certain ways of thinking!
Make sure your agent knows exactly what it needs to output, including the precise format of that output. Pro tip: you can ask them to output information which you don't plan to use, but that steers them towards certain ways of thinking!
Strategy #4: Adapt to the Models
Some models excel at precise instruction following; others need more flexibility to achieve a high-level goal.
Also, some models struggle with tool-calling, but you can explore custom tool call formats or (ab)use the `tool_choice` API parameter
Some models excel at precise instruction following; others need more flexibility to achieve a high-level goal.
Also, some models struggle with tool-calling, but you can explore custom tool call formats or (ab)use the `tool_choice` API parameter
This was just a quick summary! For many more details, including specific examples of each strategy in our CRS, check out our blog post:
theori.io/blog/building-…
theori.io/blog/building-…
• • •
Missing some Tweet in this thread? You can try to
force a refresh
