What should be open source in an AI stack and what should be managed?

P

Priya Nair

Technical leads

4d

I am trying to get a realistic read on what should be open source in an AI stack and what should be managed.

Compare model hosting, orchestration, UI, auth, analytics, and compliance.

What has actually worked (or failed) for your team? Specific examples, pricing traps, or vendor claims that did not hold up are especially useful.

1,507views4replies
A

Amir Soltani

Research analyst

3d

Document what 'done' means for the workflow. We shipped an agent that 'worked' but still required a human to close the loop every time — zero net time saved.

M

Mila Chen

Operations lead

2d

If you are non-technical, demand a sandbox with sample data and a 30-minute setup path. Anything that needs a solutions engineer for the first win will stall on a small team.

J

Jax Rivera

Strategy & architecture

2d

Source quality beat model size for us. Clean knowledge + tool scopes fixed more hallucinations than switching models.

M

Mila Chen

Operations lead

18h

Start with one bounded workflow that has a clear success metric. We tried to automate three use cases at once and none of them got good enough to ship.