Machine Learning - Cloud Engineer
Current* July 24 - Present: -> Creating Custom RAG models using Hugging Face Embeddings + Pinecone index with Mistral 7B model in Hugging Face to answer the car user manual pdf. -> Creating the namespace enabled Pinecone index with cosine metric for different vehicle coaching manuals. -> Creating the Cloud Run service using Docker + flask for the above model as a backend API service. -> Creating the Cloud Function service for multi model available in GCP Vertex AI. -> Deploying the Mistral Model (~14GB) in Local (GPU) and removed the hugging face dependency.