Build a Cloud-Native RAG-Powered Knowledge Assistant (Gen-AI, LangChain, Vertex AI/AWS Bedrock)

We're looking for a highly skilled Gen-AI Engineer to build a mini production-ready Knowledge Assistant using RAG (Retrieval-Augmented Generation) architecture. The assistant should ingest a small set of documents (PDFs or text), store them in a vector database, and allow query-based interactions using LLMs via Google Vertex AI / AWS Bedrock / Gemini API.
This is a small project (10–20 hours) aimed at showcasing your ability to build intelligent, cloud-native Gen-AI applications.
______________________________________
Deliverables:
•⁠ ⁠Set up cloud environment (GCP/AWS)
•⁠ ⁠Ingest documents and build a vector store (e.g., FAISS, Pinecone, or similar)
•⁠ ⁠Implement a basic RAG pipeline using LangChain or LangGraph
•⁠ ⁠Expose the system via FastAPI (or Django minimal setup)
•⁠ ⁠Deploy the application using Docker and provide a working CI/CD flow (GitHub Actions or similar)
•⁠ ⁠Basic UI or API documentation for testing queries
______________________________________
Preferred Tech Stack:
•⁠ ⁠LLM Platform: Google Vertex AI / AWS Bedrock / Gemini API
•⁠ ⁠Framework: LangChain / LangGraph
•⁠ ⁠Backend: FastAPI / Django
•⁠ ⁠Vector Store: FAISS / ChromaDB / Pinecone
•⁠ ⁠Infrastructure: GCP / AWS
•⁠ ⁠Containerization: Docker
•⁠ ⁠CI/CD: GitHub Actions / Cloud-native pipeline
______________________________________
Ideal For:
•⁠ ⁠Gen-AI engineers seeking to build review-worthy, demonstrable projects
•⁠ ⁠Professionals wanting to showcase LLM orchestration, RAG, and deployment skills
•⁠ ⁠Developers experienced in AI automation and enterprise use-case alignment
______________________________________
Timeline:
•⁠ ⁠Estimated Time: 10–20 hours
•⁠ ⁠Delivery: Within 1 week preferred