GreenOps Architect: AI Cloud Validator
AI Agent & Cloud Architecture
Year
2026
Context
Final Project: IT - AI Agent for Programming (Hacktiv8 × IBM SkillsBuild)
Role
AI Solutions Architect
Gemini 2.5 Flash
Astra DBProject Overview
An enterprise-grade AI agent powered by Langflow and Gemini that automatically audits proposed cloud architectures to prevent financial waste and reduce carbon emissions.
The Problem
Engineering teams frequently over-provision cloud infrastructure to 'play it safe,' leading to massive AWS billing bloat and a severe environmental carbon footprint. Furthermore, manually auditing proposed architectures against dense AWS best-practice whitepapers is highly time-consuming and prone to human error.
The Solution
I engineered a custom Vector Store RAG pipeline using Langflow, Astra DB, and Google Generative AI. To bypass rate limits, I built a batch ingestion pipeline that chunked a 135-page AWS Cost Optimization Whitepaper into sequential micro-batches. I then injected an 'Elite Cloud Architect' persona to rigorously analyze user prompts against the vectorized data.
The Impact & Learning
Delivered a dual-impact AI auditor for DevOps and FinOps teams. The agent analyzes architecture proposals in real-time and generates structured Markdown reports highlighting cost risks, sustainability impacts, and actionable right-sizing recommendations (e.g., adopting Serverless or AWS Graviton) before a single line of infrastructure code is written.
Live Demonstration