How I Stopped Being an “Old-school Dev” to Create an Artificial Intelligence Monitor with Gemini

2026-06-21 · Source: Data Engineering on Medium · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering · Depth: Intermediate, short

Summary

A developer, initially resistant to AI, transitioned to AI-assisted development, significantly boosting productivity. This shift, however, highlighted the unmanaged costs associated with LLM API token consumption across personal projects. To address this, the developer created the "Token Monitor," a lightweight, 100% local application. Developed with Gemini and the SuperPowers extension, this tool parses project directories to centralize token usage, offering temporal mapping, granularity by LLM model (e.g., Gemini, GPT, Claude), project/model filters, and a dynamic pricing layer to translate token counts into real financial costs. The project's code has been published on GitHub, demonstrating a practical solution for managing LLM expenses.

Key takeaway

For AI Engineers or MLOps teams managing LLM-driven projects, proactively monitoring token consumption is critical to control costs and optimize model usage. Your projects, even personal ones, accrue significant API expenses that can quickly become unmanageable without visibility. Consider developing or adopting local monitoring solutions to gain granular insight into API costs across different models and projects, ensuring financial sustainability and informed decision-making.

Key insights

Embracing AI-assisted development necessitates proactive token consumption monitoring to manage LLM API costs effectively.

Principles

• Prioritize delivery value over code purism.
• AI tools increase productivity but incur token costs.
• Monitor LLM API usage for financial sustainability.

Method

Build a local application to parse project directories, read consumed tokens, and visualize usage with temporal mapping, model granularity, project filters, and dynamic pricing layers.

In practice

• Use AI for code validation, refactoring, and pair programming.
• Monitor LLM API costs with a local, custom-built tool.
• Publish useful internal tools to GitHub for community benefit.

Topics

• LLM Cost Management
• AI-assisted Development
• Token Monitoring
• Gemini API
• Developer Productivity
• MLOps Observability

Code references

• abnerfc01/token-monitor

Best for: AI Engineer, Machine Learning Engineer, MLOps Engineer

Related on AIssential

• Token Spend Out of Control? The Case for Smarter Routing — Dynamically routing LLM agent requests to the cheapest capable model is crucial for controlling escalating token spend.
• You can't trust one token number across your tools. Here's the guide to a dashboard that keeps Codex, Claude, and ChatGPT honest. — A token burn dashboard reveals AI usage habits, improving outcomes and expanding imaginative computing beyond mere cost tracking.
• Tracking Every Token: Granular Cost and Usage Metrics for Microsoft Foundry Agents — Granular AI agent cost and usage tracking is achievable by integrating Azure API Management and Application Insights.
• Meta shifts from "tokenmaxxing" to token managing as internal AI costs reportedly hit billions — Unmanaged internal AI usage can lead to billions in costs and artificial consumption without productivity gains.
• Nobody Told You AI Would Cost This Much in Production. Here’s the Real Invoice. — LLM production costs extend far beyond token pricing, driven by infrastructure, retries, and inefficient prompt management.
• Your LLM bill is not your infra bill: a budgeting catalog for AI-feature SaaS — AI costs are a metered utility, not fixed infrastructure, requiring proactive behavioral budgeting to prevent unexpected spikes.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Data Engineering on Medium.