Most technical writers don't build AI agents. We document them. We sit in on the sprint review. We read the architecture decision records. Then we figure out how to explain a retrieval pipeline to someone who needs their integration working by Friday.

I built one anyway.

His name is Artie, an AI documentation assistant I built with Algolia. He runs inside this documentation site. Building him taught me more about writing and AI than any conference talk could.

This isn't a walkthrough of Artie's design. The case study covers that. This post covers the broader lessons:

• What AI agents actually are
• How retrieval-augmented generation works in practice
• Why prompt engineering is closer to technical writing than most people realize
• What this all means for people in this little thing of ours

"What is New York if not 99.6 percent of the world stacked against you? And who are New Yorkers if not people who hear those odds and smile? [...] It is in that 0.4 percent that we go to work."

— Mayor Zohran Kwame Mamdani

Well, it's been a whole summer since that faithful night in June when the Knicks fulfilled the dream generations of New Yorkers shared for 53 years. The city is still standing. The whole world thought New York City would explode, or implode, or something from the sheer chaos that would ensue, but we're still here.

Instead, the city exploded in love, one harmonious celebration echoing through every corner of the towns.

THE NEW YORK KNICKS ARE THE BEST IN THE WORLD! THE LARRY IS COMING HOME TO THE GARDEN.

Finally, the phrase "living through unprecedented times" means a good thing.

I love you, Jalen Brunson. ¡Big KAT, de lo mío personal ! Josh Hart—the human swiss army knife—you're the new Excellence of Execution. Landry Shamet, we dub thee "Champagne Shamet." Big Mitch, you've seen this team go from zeroes to heroes.

Mikail Bridges, who might be a Viltrumite? You've shown us what it truly means to be—