How to Scale an LLM Ops Pipeline

I think it goes back to my framework is that people should be realistic what can they achieve and start with your skills. Even on those, you should focus on easier use cases because many use cases, latency is less of an issue. So the first and foremost, I think classical role focused on use cases that are achievable. And then even on those use cases, ton of work because you need to make sure also even the point was, you know, open-ended NLU-based input. It sounds very cool. Oh, let me allow people to say whatever they want. But what I discovered in my many years for example, speech language is extremely rich,. And it's awesome,

Play episode from 31:06

Transcript

Episode notes

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!

Get the app