Telegram bot — AI aggregator with monetization

Created a Telegram bot unifying multiple AI models in one interface. Implemented a credit and billing system, admin panel for managing models and users, and a scalable architecture with task queues. The project reached monetization within the first month.

The client wanted to offer multiple AI models (GPT-4, Claude, Midjourney, Stable Diffusion) through a single Telegram bot with pay-per-use monetization. The challenge: handling concurrent AI requests without blocking, managing user credits, and keeping infrastructure costs predictable.

• Built an async architecture with Celery + Redis for processing AI requests in parallel
• Implemented a credit system with multiple top-up options (Stripe, crypto, in-app purchase)
• Created an admin panel for managing AI model configurations, pricing, and user accounts
• Added queue prioritization — paid users get faster processing
• Built cost tracking per model to maintain margins on each API call

• 500+ paying users within 3 months of launch
• Monetization achieved in month 1 — revenue covered infrastructure costs
• 99.7% uptime with graceful handling of AI API rate limits and errors
• Average 2,000 AI requests/day processed across all models
• 30% of free users converted to paid plans