Pre-screened and vetted.
“Built and deployed an LLM-powered financial document processing and summarization platform at Morgan Stanley using a production RAG pipeline (PDF ingestion, embedding-based retrieval, schema-constrained JSON outputs) delivered via FastAPI microservices on Kubernetes. Drove measurable impact (40% reduction in manual review time) and improved factual accuracy for numeric fields by 30% through metadata-aware retrieval, strict schemas, and post-generation validation, with a human feedback loop from financial analysts.”
“Built an automated ML/NLP document classification system for unstructured legal documents, combining classical models (TF-IDF + logistic regression/random forest) with entity resolution via fuzzy matching validated by precision/recall. Also implemented semantic similarity search using sentence embeddings stored in FAISS and improved matching by fine-tuning a transformer on domain-specific data and tuning similarity thresholds for fewer false positives.”