Hi, quick note before anything else  this kind of tool usually looks simple on the surface, but it’s exactly the type that quietly defines whether a business scales smoothly or keeps fighting its own systems.

I’ve worked on a similar internal content engine for a services company where the real value came from letting the system learn at the client level instead of averages but are you imagining feedback as explicit ratings, edits, or passive signals like usage and revisions? 

Since you want to stay with Claude, I’d focus on wrapping it in a clean, modular environment that keeps prompting flexible while allowing future model swaps also are you planning for multiple teams or just internal use at first?

For translations and improvement over time, I’d suggest a separate core intent generation from language rendering so learning doesn’t get muddled but  do you need consistent tone per client across languages? 

And when you say long-term, are you thinking versioned prompts, client memory, or full auditability for outputs? On the infrastructure side, I’d suggest a scalable but simple setup so you’re not boxed in later and would be nice if like are you deploying this internally, or does it need client-facing access?

Sorry for asking way too much of questions but clarity brings every project to obtain the right success! 


Thanks. 

Based on scope and durability, this feels more like a foundational build than a quick tool, so I’d expect a budget in the £2.8k–£3.5k range depending on how deep the feedback system goes. If it helps, I can walk you through a clean system design on a short call so you can sense-check whether the approach fits how you want to grow this — would that be useful right now?