Optimize AI Model Token Efficiency

I am seeking an experienced developer to help optimize the AI model usage in my n8n project, which processes a large quantity of time series data. The current setup involves sending multiple spreadsheets, merged and aggregated, to an AI Chat Model using GPT-4.1-Mini, consuming 295,000 tokens per execution. The data updates every 15 minutes, but much of it remains unchanged between queries. My primary goal is to minimize token usage while still utilizing all previous data effectively.

Key Requirements:
- Analyze and optimize the current query structure to reduce token consumption.
- Implement token-efficient encoding methods to enhance the efficiency of data processing.
- Ensure that the AI model usage remains effective with minimized token usage.
- Maintain the integrity and accuracy of data processing and AI interactions.

Ideal Skills and Experience:
- Strong understanding of AI models, particularly GPT-4.1-Mini.
- Experience with token optimization and efficient data encoding methods.
- Proficiency in workflow automation tools like n8n.
- Ability to analyze and restructure data queries for improved efficiency.

If you have the expertise to optimize AI model usage and reduce token consumption effectively, I would love to hear from you.