AI as an API in Real-World Applications
Annotation
Most of us have tried ChatGPT or assistants like GitHub Copilot. But what happens when you take an AI API and make it the core of your product? Using the real-world example of Contember AI Studio, I’ll walk you through what to expect and what challenges you’ll need to tackle. We’ll demonstrate everything using real-life stories (and plenty of dead ends) from our daily work over the past few days and months.
You’ll learn:
- How much you can rely on the output from an AI API
- What tooling you need if you’re serious about your application
- How—and if—AI agents for code editing work
- And (if nothing goes too wrong), how far you can go with fine-tuning GPT-3.5