Projects

Resume Parsing System

Built an NLP pipeline to extract structured data from resumes using spaCy, HuggingFace Transformers and DeBERTa-v3, deployed via FastAPI.

PythonHuggingFace TransformersspaCyBERTDeBERTa-v3FastAPIDocker

Resume–JD Matching Engine

Contextual similarity scoring using DeBERTa embeddings instead of keyword matching.

DeBERTaTransformersCosine Similarity

LLM Resume Analyzer

LLM-powered feedback system that analyzes resumes and suggests improvements.

LLMsPrompt EngineeringFastAPI

Async ML Processing Pipeline

Handled concurrent ML tasks using Celery and Redis for scalable inference.

CeleryRedisDocker