Questions to ask
- 1. Will you build it, or just recommend it? Advice without execution usually ends in a folder.
- 2. Can you show me something you built that is running today?
- 3. How do you decide what to automate first?
- 4. Who owns the accounts, code and data? The answer should be you.
- 5. How do you handle mistakes and accuracy? Look for testing, guardrails and optional human review.
- 6. What happens after launch? AI systems need monitoring and updates.
- 7. Will this work with the tools we already use?
- 8. How will we measure whether it worked?
Red flags
- Promises of specific results before they understand your business.
- Locking you into their own platform or accounts.
- No plan for maintenance or for what happens when something changes.
- Lots of jargon and no plain-language explanation of how it will work.
What good looks like
A good AI partner starts small, proves value on one real process, and earns the right to do more. They explain things simply, test on your real data, and leave you owning everything.
Related: AI consulting and AI consulting in the Bay Area.