My Lovable account is burning through credits faster than expected, mainly during data-processing steps. I want to tighten things up so every AI task runs as efficiently as possible without losing quality. Here’s what I need from you: • Audit current usage: walk through my existing Lovable workflows, calls, and settings to pinpoint exactly where credits are being spent. • Optimisation plan: deliver clear, step-by-step recommendations—batching, caching, parameter tweaks, or architectural changes—that specifically reduce the cost of data processing while maintaining output accuracy. • Proof-of-concept implementation: update one or two representative tasks so we can benchmark before-and-after credit consumption. Share concise code snippets or configuration files (Python preferred, but any language that works with the Lovable API is fine). • Reporting dashboard: set up a lightweight monitor or script that tracks credit usage per task going forward so I can see the savings in real time. Acceptance criteria: a documented reduction in credits per processed dataset and a reusable workflow I can extend to the rest of the account. If your background includes deep familiarity with Lovable’s API limits, token accounting, and data-handling best practices, let’s talk—this is exactly your wheelhouse. My goal is to: Optimize my current project to reduce credit consumption Improve prompt structure for efficiency Implement best practices to avoid unnecessary usage Possibly redesign parts of the workflow to lower overall costs I am NOT looking for account manipulation or unofficial credit purchasing. Only legal and technical optimization solutions. If you have proven experience working with Lovable or similar AI-based platforms, please share your experience and approach.