An end-to-end AI-powered pipeline that ingests academic papers from arXiv, extracts and validates text, generates structured AI summaries via GPT-4.1-mini, computes vector embeddings, and serves results through a hybrid semantic search API backed by Azure AI Search — with real-time monitoring via Prometheus and Grafana.
PythonFastAPIStreamlitAzure Blob StorageAzure AI SearchOpenAI APISentence TransformersDockerTerraformPrometheusGrafana