Can I vibe code Perpetua?

Build a light retail media ad optimization service in Python and FastAPI for Amazon Advertising API. 1. DATA MODEL: Create PostgreSQL tables for Accounts (client_id, refresh_token), Campaigns (campaign_id, platform, budget, target_acos), Keywords (keyword_id, campaign_id, text, match_type, bid, sales, spend, impressions, clicks), and SearchTermLogs (query, conversions, spend). 2. API INTEGRATION: Implement OAuth 2.0 refresh flow for Amazon Advertising API. Add daily CRON tasks to request, download, and parse Sponsored Products Performance and Search Term reports. 3. BIDDING ENGINE: Write a rule-based bidding script: Calculate current ACoS = (spend / sales). If sales > 0 and ACoS > target_acos, decrease bid by 10% (minimum bid $0.20). If sales > 0 and ACoS < target_acos, increase bid by 10% (maximum bid $5.00). If clicks > 15 and sales == 0, decrease bid by 20%. 4. KEYWORD HARVESTING: If a search term in SearchTermLogs has >= 2 conversions and is not already a target, create an Exact match keyword in the campaign and add a Negative Exact entry to the Auto campaign. 5. FAILURE MODES: Handle API rate limiting (HTTP 429) using exponential backoff. Account for Amazon's 14-day attribution lag by evaluating metrics over a trailing 30-day window excluding the last 48 hours. Out of scope: Machine learning bid forecasting, Walmart/Instacart integrations, and complex AMC SQL queries.

$ each button prefixes agent-specific run instructions · build your own product, never copy proprietary code, trademarks or designs