Skip to content

Ask HN: Should lack of ML expertise affect scaling ideas based on LLMs?

1 pointsanmon3186discuss
On HN

Coming from traditional software dev background in enterprise IT, I am fascinated by what foundation LLMs can do.

I don't have a deep background in ML, fine-tuning, training models etc, but I have experimented with plain OpenAI APIs with python/LangChain tooling to build few small apps.

What pulls me back beyond a point is relying solely on these models, thinking it might limit differentiation in the market without core AI capabilities and cost-benefit calculations for tons of API calls vis-a-vis homegrown models.

May be it’s just my insecurities but I'm curious to hear your thoughts and experiences.

Comments

No comments yet.

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.