Feasibility is rarely a flat no. It is a price. The question for engineers is not whether this can be built — almost anything can — but what it would cost.
Get them looking while the work is still rough: sketches, not finished screens. A two-day spike answering one technical question beats a fortnight of estimating. Ask which part they are least sure about; the uncertainty is where the schedule breaks.
Cost is not only technical. Ask what else is in flight before you ask how long — a buildable design is still out of reach if the two engineers who know that service are committed elsewhere.
Then sort what you hear into three kinds of infeasible. Impossible on this architecture — change the design, or fund the rebuild. Possible but expensive — a trade: name what it displaces and let whoever owns the roadmap choose. Possible but slow — start now, or cut what waits on it.
An estimate from someone who has not seen the design measures their optimism, not the work.