Skip to content
Back to Projects
ML / AI 2024-03

RAG Support Assistant

Internal RAG-powered support assistant with Slack integration, OpenAI, and ChromaDB. Hosted on CVS Enterprise GCP via Terraform + ArgoCD; reduced manual ticket resolution time by 20%.

20% Faster Resolution
RAGOpenAIChromaDBSlackTerraformArgoCDGCPLLMPython

Overview

Built an internal Retrieval-Augmented Generation (RAG) support assistant at CVS Health that integrates with Slack to provide self-serve troubleshooting for engineering teams. The system uses OpenAI for generation and ChromaDB for vector storage, with secured governance, and is hosted on CVS Enterprise GCP clusters via Terraform and ArgoCD.

Key Features

  • Slack Integration: Native Slack bot for seamless team interaction
  • RAG Architecture: ChromaDB vector store for document retrieval, OpenAI for answer generation
  • Self-Serve Troubleshooting: Teams can resolve common issues without filing tickets
  • Context-Aware Responses: Retrieves relevant documentation and past solutions
  • Secured Governance: Enterprise auth, request audit logs, and content guardrails

Impact

  • 20% reduction in manual ticket resolution time
  • Accelerated onboarding for new team members
  • Reduced load on platform support teams
  • Improved knowledge discoverability across the engineering organization

Tech Stack

  • AI/ML: OpenAI API, ChromaDB (vector database), RAG pipeline
  • Integration: Slack Bot API
  • Backend: Python, FastAPI
  • Infrastructure: GCP (CVS Enterprise clusters), Terraform, ArgoCD
  • Data: Internal documentation corpus, runbook collection

Related Projects

ML / AI

AI-Driven Image-to-Alt-Text CI Automation

Hugging Face transformer-powered image alt-text compliance system at CVS Health. Won 1st Prize at the CVS GenAI Hackathon (Aug 2023), then productionized for enterprise rollout — boosted accessibility conformance by 35% and reduced downstream defects by 15%.
ML / AI

CVS GenAI Hackathon — Alt-Text Compliance

Awarded 1st Prize at the CVS GenAI Hackathon (Aug 2023) for a proof-of-concept ML transformer application that generates and compares image alt-text and tests it for accessibility compliance. Productionized into the CVS image-to-alt-text CI pipeline.