Agent Building is Complex Engineering: Building agents is a new field, but good AI engineering follows the principles of good software engineering.
Evaluating agents - Blog - Braintrust
When building applications with LLMs, we recommend finding the simplest solution possible, and only increasing complexity when needed.
Start by using LLM APIs directly: many patterns can be implemented in a few lines of code. If you do use a framework, ensure you understand the underlying code.
anthropic-cookbook/patterns/agents at main · anthropics/anthropic-cookbook
Once the task is clear, agents plan and operate independently, potentially returning to the human for further information or judgement.
The autonomous nature of agents means higher costs, and the potential for compounding errors.
Success in the LLM space isn't about building the most sophisticated system. It's about building the right system for your needs. Start with simple prompts, optimize them with comprehensive evaluation, and add multi-step agentic systems only when simpler solutions fall short.
Workflow follows a fixed set of steps in a strict sequence, relying on predefined prompts and decisions.
Agent is more open ended and adaptive, determining how many iterations it needs.
Agents can handle sophisticated tasks, but their implementation is often straightforward. They are typically just LLMs using tools based on environmental feedback in a loop.