What I’m finding interesting are the friction points in these workflows: • Copy-paste between Perplexity and your IDE • Manually deciding which model to use for each task • Context switching between different interfaces • No learning/optimization over time
The feedback is making me reconsider Maurice’s positioning. Rather than “90% cheaper,” maybe the real value is “eliminate the manual orchestration you’re already doing.”
A few questions for you (and others doing similar optimization): 1. How much time do you spend on the copy-paste workflow daily? 2. Do you find yourself wishing for automatic model selection based on task complexity? 3. Would seamless context preservation across providers be valuable?
This is exactly why I posted — assumptions about “cheaper” vs “more automated” need real validation. Appreciate the reality check.
I’ve been spending more than $200–$400/month on AI services (Claude, GPT-4, Cursor, etc.) for development work, and I’m starting to wonder if I’m not alone in feeling the pinch.Questions for the community: 1. What’s your monthly AI spend? (Including API costs, subscriptions, etc.) 2. Have you changed behavior due to costs? • Reduced usage of premium models? • Switched to cheaper...