Your point that intent is ambiguous on purpose rings true in government. Policy intent often arrives as the wording everyone could agree to, and delivery teams have always quietly made the missing decisions. In public services, user research, content design and interaction design are where much of that intent gets made clear. If agents act on ambiguity instead of absorbing it, those disciplines matter more, but they need to work closer to the build.
Your intent-and-verification loop matched my own experience too. With a coding agent, shared documentation of decisions mattered more than the speed of the code.
Thanks Mark — and welcome. Sadly it's not a public-sector speciality — I've seen the same pattern with plenty of enterprise customers. The wording everyone can agree to is usually the wording that decides nothing.
And you're right that user research and content design matter more now — and luckily they've never been easier to do close to the build.
One thing I'd add: clarifying intent early is still far cheaper than discovering it in delivery — but it needs a loop back to whoever originally asked: is this still what you meant? That check used to happen implicitly, courtesy of the slowness. Now it has to be designed in.
And fully agree on shared decisions. In platform engineering I always recommended publishing ADRs openly, so intent providers and executors work from the same record. With agents, that record becomes their context too.
Curious: what did your decision documentation look like with the coding agent — a decision log, ADRs, something else?
Your point that intent is ambiguous on purpose rings true in government. Policy intent often arrives as the wording everyone could agree to, and delivery teams have always quietly made the missing decisions. In public services, user research, content design and interaction design are where much of that intent gets made clear. If agents act on ambiguity instead of absorbing it, those disciplines matter more, but they need to work closer to the build.
Your intent-and-verification loop matched my own experience too. With a coding agent, shared documentation of decisions mattered more than the speed of the code.
Thanks Mark — and welcome. Sadly it's not a public-sector speciality — I've seen the same pattern with plenty of enterprise customers. The wording everyone can agree to is usually the wording that decides nothing.
And you're right that user research and content design matter more now — and luckily they've never been easier to do close to the build.
One thing I'd add: clarifying intent early is still far cheaper than discovering it in delivery — but it needs a loop back to whoever originally asked: is this still what you meant? That check used to happen implicitly, courtesy of the slowness. Now it has to be designed in.
And fully agree on shared decisions. In platform engineering I always recommended publishing ADRs openly, so intent providers and executors work from the same record. With agents, that record becomes their context too.
Curious: what did your decision documentation look like with the coding agent — a decision log, ADRs, something else?