I’m a Machine Learning Engineer focused on NLP, LLMs, and Transformers(more classical approach).
My work blends engineering with research-driven modeling — from hierarchical transformers to LLM-assisted dataset generation, entity extraction, and large-scale text processing .
I love building systems that understand messy real-world text, uncover structure, and power analytics, recommendations, and search.
🔭 Current deep-dive:
Designing a context-aware transformer that fuses sentence-level, section-level, and document-level reasoning for robust employer/entity extraction.
A multi-level transformer architecture that integrates:
• Sentence context
• Section-level signals (COMPANY, SKILLS, JOB, etc.)
• Document-wide TF-IDF–guided pruning
Designed for accurate entity extraction + employer detection.
Tech: PyTorch, DistilBERT, attention layers, TF-IDF scoring
Local retrieval agent using
LlamaIndex + LangChain + Ollama models
for PDF question answering and contextual summarization.
Combined content-based & collaborative filtering
for skincare recommendations with a simple Flask/Streamlit UI.
Built an optical flow + LSTM system to detect heavy-object anomalies
in real-time waste-sorting conveyor footage.
✨ Always open to collaborating on ML/NLP research, transformer architectures, and meaningful AI projects ✨