The problem, the architecture, and the number that moved.
Engagements written up by the engineers who delivered them. Client names are withheld where the work sits under an NDA — we will walk you through the detail on a call.
3 of 3 engagements
[Headline: the change the client got, in one line]
[Two sentences: where the platform stood before, what you built, and the outcome the client measured.]
Read the engagement[before → after]
[what this measures]
[N%]
[what improved]
[N weeks]
[time to first result]
[Headline: what the assistant or model does for them now]
[Two sentences: the manual work it replaced, how you made it trustworthy, and what changed after launch.]
Read the engagement[N%]
[what improved, e.g. handling time]
[N pts]
[what moved, e.g. resolution rate]
[100%]
[the guarantee you can make]
[Headline: the decision the model now supports]
[Two sentences: how the decision was made before, what you built instead, and the number that moved.]
Read the engagement[−N%]
[what shrank, e.g. buffer or cost]
[N pts]
[accuracy improvement]
[Weekly]
[the review that now exists]
Project impact and return on investment.
Every engagement in the numbers the client measured after go-live: what the work returned, how long it took, and which metric moved.
- Return on investment
[Nx return]
Duration[N weeks]
Key metrics- [before → after][what this measures]
- [N%][what improved]
- [N weeks][time to first result]
- Return on investment
[Nx return]
Duration[N weeks to pilot, then ongoing]
Key metrics- [N%][what improved, e.g. handling time]
- [N pts][what moved, e.g. resolution rate]
- [100%][the guarantee you can make]
- Return on investment
[Nx return]
Duration[N weeks]
Key metrics- [−N%][what shrank, e.g. buffer or cost]
- [N pts][accuracy improvement]
- [Weekly][the review that now exists]
Bracketed figures are placeholders. Each row reads its client, return, duration, metrics and outcome from the matching case study entry.
