AI DevOps Project #2 - AI Cloud Cost Detective (Reduces Cloud Billing)
Why It Matters
By automating real‑time cost‑optimization recommendations, the AI Cloud Cost Detective helps enterprises prevent overspend before it occurs, turning cloud budgeting from a reactive to a proactive discipline.
Key Takeaways
- •AI tool scans cloud resources to suggest cost‑saving changes.
- •Built with open‑source stack: React, FastAPI, PostgreSQL, Azure CLI.
- •Uses LLM (OpenAI) to compare configurations against best practices.
- •Architecture includes authentication, request flow, and investigation history storage.
- •Project is cloud‑agnostic; replace Azure commands for AWS or GCP.
Summary
The video walks viewers through creating an "AI Cloud Cost Detective," a custom AI‑driven utility that connects to a cloud provider, inventories resources, and recommends configuration tweaks to curb overspending. Abhishing outlines a five‑step roadmap—problem definition, tooling, architecture, request flow, and hands‑on implementation—using open‑source components such as React for the UI, FastAPI in Python for the backend, PostgreSQL for data and auth, and Azure CLI for resource discovery. The core intelligence relies on a large language model (e.g., OpenAI) that benchmarks current settings against documented best‑practice configurations, then returns actionable summaries and command snippets for VMs, S3 buckets, Kubernetes clusters, and more. He emphasizes enterprise‑grade features like user authentication, persistent investigation history, and a modular design that can be swapped to AWS or GCP by adjusting prompts and CLI calls. The project demonstrates how DevOps teams can shift from reactive monthly cost reports to proactive, AI‑guided optimization, potentially saving tens of thousands of dollars per month.
Key insights include the prevalence of mis‑configured or over‑provisioned resources driving unexpected cloud bills, the limitations of existing open‑source tools that target specific services, and the advantage of a unified AI layer that can span any resource type. By leveraging Azure CLI’s stability over volatile APIs, the solution minimizes maintenance overhead while ensuring compatibility across cloud generations. The architecture places the AI model behind a secure FastAPI endpoint, stores user and analysis data in PostgreSQL, and presents results through a React dashboard, creating a full‑stack, self‑hosted platform.
Abhishing highlights that "instead of waiting till the end of the month and paying $50,000, organizations can reduce that to $20,000" using the detective. He also notes that the repository includes detailed diagrams and request‑flow documentation, which improve prompt accuracy for AI coding assistants like GitHub Copilot or Cursor, reducing the risk of implementation errors.
For businesses, the tool offers a proactive cost‑control mechanism, turning cloud spend from a surprise expense into a manageable metric. Its extensibility across cloud providers and reliance on open‑source stacks make it an attractive, low‑cost alternative to proprietary cloud‑cost management suites.
Comments
Want to join the conversation?
Loading comments...