The AI Was Never the Hard Part

Description and use

<p>Twenty-four articles. One agent. A lot of things I got wrong before I got them right.</p> <p>This one isn't about code. It's about what I'd tell myself at article 1 if I could go back.</p> <h2> I thought the model was the hard part </h2> <p>It isn't. Spring AI handles the model. You call <code>chatClient.prompt().user(message).call().content()</code> and it works. Five lines.</p> <p>The hard part is everything around that call. How do you pass context without spending a fortune on tokens? What happens when a tool fails mid-conversation? How do you know if the response is actually correct? H

Problem

<p>Twenty-four articles. One agent. A lot of things I got wrong before I got them right.</p> <p>This one isn't about code. It's about what I'd tell myself at article 1 if I could go back.</p> <h2> I thought the model was the hard part </h2> <p>It isn't. Spring AI handles the model. You call <code>chatClient.prompt().user(message).call().content()</code> and it works. Five lines.</p> <p>The hard part is everything around that call. How do you pass context without spending a fortune on tokens? What happens when a tool fails mid-conversation? How do you know if the response is actually correct? H

Outcome

<p>Twenty-four articles. One agent. A lot of things I got wrong before I got them right.</p> <p>This one isn't about code. It's about what I'd tell myself at article 1 if I could go back.</p> <h2> I thought the model was the hard part </h2> <p>It isn't. Spring AI handles the model. You call <code>chatClient.prompt().user(message).call().content()</code> and it works. Five lines.</p> <p>The hard part is everything around that call. How do you pass context without spending a fortune on tokens? What happens when a tool fails mid-conversation? How do you know if the response is actually correct? H

Evidence

The AI Was Never the Hard Part

Source