Is the idea ready
for your roadmap?
A demo can make an AI feature look convincing. Before your team commits to building it, try it against the customer’s actual job and the cases most likely to cause trouble.
Name the customer task.
What is the customer trying to finish? What do they do today, and what happens if the feature gets something wrong? Gather examples your team can return to throughout testing.
Agree what would count as useful before comparing approaches. That gives product and engineering the same basis for a decision.
Build enough to compare.
Try a small number of designs. The job may call for a model, document retrieval, computer vision, ordinary software or a person reviewing the output.
Keep the inputs and configurations alongside the results. When one version looks better, you should be able to repeat the test and explain why.
Look beyond the successful demo.
Check errors, response time, running cost and the effort needed to connect the feature to your product. Count the time people spend reviewing or correcting its answers, too.
Then decide whether to pilot it, narrow it, keep investigating or set it aside. I help your team make that decision with a record of what you tried and what you found.
We can work through this together.
This is an example of a project we could explore in the 90-day programme. Your team builds it, with my teaching and review along the way. We use Intent-to-System Engineering to keep the tests connected to the job.
Tell me what you’re working on.
A few lines about the problem and your team are enough to start. We can talk through the rest together.