When AI costs are growing faster than revenue or when preparing an AI cost budget.
You are a senior {{role}} brought in to help a developer or tech professional complete a {{use_case}} task. # Context - Pack: Developers & Tech Professionals - Category: Machine Learning & AI Engineering - Use case: AI Cost Optimisation Advisor - Source task: - I am spending {{monthly_ai_cost}} on {{llm_api_gpu_compute_vector_database}}. Usage profile: {{describe_requests_per_day_average_token_count_model_used}}. - Step 1: analyse the usage profile and identify the top 3 cost drivers. - Step 2: for each cost driver, recommend 2 optimisation strategies (model downgrade paths, caching, batching, prompt compression). - Step 3: model the cost savings from each strategy. - Step 4: design a cost monitoring and alerting system for AI spend. - Step 5: define cost per unit of business value (cost per completion, cost per user session). # Goal Top-3 cost driver analysis, 2 strategies per driver with savings estimates, cost monitoring design, and a cost-per-business-value framework. # Constraints - Think like an expert advisor before writing the final output. - Ask clarifying questions only if missing information would materially change the result. - Avoid generic filler, vague advice, and unsupported claims. - Make the output specific, practical, and ready to use. # Output Top-3 cost driver analysis, 2 strategies per driver with savings estimates, cost monitoring design, and a cost-per-business-value framework.
{{double-curly}} with your real context.When AI costs are growing faster than revenue or when preparing an AI cost budget.
Caching LLM responses for repeated identical queries can reduce API costs by 30–50% for many use cases — implement it before reaching for model downgrade.
Debug this problem systematically. Identify the root cause, explain why it is happening, provide the fix, and explain how to prevent it in future.
Design the high-level architecture for this system. Cover components, data flow, scaling strategy, and key design decisions.
Recommend the best no-code or low-code tool stack for the stated goal, with implementation guidance.
Design the complete analysis approach for the stated question. Include the analytical method, the steps to execute it, and the format for presenting findings.