I'm a Lead ML Engineer with a passion for architecting and productionizing cutting-edge AI solutions, especially in Generative AI, LLMs, NLP, and Computer Vision. I thrive on solving complex problems and driving innovation from proof-of-concept to deployment.

Currently, I'm also pursuing a Ph.D. in Data Science and Artificial Intelligence at IIIT Bangalore, focusing on Advanced LLM applications, multimodal LLMs.

• Leading ML initiatives at Zycus Infotech, focusing on LLM pipelines, RAG systems, and multi-agent solutions for the procurement domain.
• Deep diving into advanced fine-tuning techniques (PEFT, LoRA, Quantization) for models like LLaMA 3.
• Building robust MLOps workflows for efficient model deployment and monitoring.
• My Ph.D. research at IIIT Bangalore.

Here's a snapshot of the technologies and tools I work with:

Core Areas:

Languages & Databases:

LLMs & GenAI Frameworks/Tools:

ML/DL Libraries:

Cloud & MLOps:

Other Tools:

(09/2023 - Present)

• Architected & Productionized end-to-end LLM pipelines (GPT-4/3, LangChain) for automated invoice data extraction (97% accuracy).
• Engineered & Deployed scalable RAG systems using Knowledge Graphs (Neo4j) and Vector DBs (Pinecone, FAISS) on AWS, improving semantic search by 20%.
• Executed advanced fine-tuning (PEFT, LoRA, Quantization) on LLaMA 3 models.
• Established robust MLOps workflows with CI/CD pipelines.
• Integrated multimodal models (LayoutLM, DocVQA), enhancing data extraction by 25%.
• Exploring around multi-agent system POC (AutoGen) for procurement.

(04/2022 - 09/2023)

• Helped in building Discharge Summary Digitizer, fine-tuning a clinical NER model (Transformers) to 94% accuracy, reducing manual review by 35%.
• Developed an innovative multimodal model (LayoutLMv2, CV) for Lab Report Digitization.
• Optimized document classification (+28%) via fine-tuning Faster R-CNN/YOLO models.

(12/2019 - 06/2020)

• Researched & Implemented a custom CNN-LSTM architecture for mental workload estimation from EEG (84% accuracy).
• Designed insightful dashboards using Dash, Tableau, and Plotly.

• Ph.D. in Data Science and Artificial Intelligence (Part-time)
  • INTERNATIONAL INSTITUTE OF INFORMATION TECHNOLOGY BANGALORE (IIITB), India
  • 2024 - Current
• Master of Science in Computer Science (Major: Data Science)
  • CHRIST UNIVERSITY, Bangalore, India
  • 2020 - 2022
• Bachelor of Computer Application (Major: Computer Science)
  • GGSIP UNIVERSITY, New Delhi, India
  • 2017 - 2020

• Sign Language Recognition System:
  • Published in: Proceedings of the International Conference on Innovative Computing & Communication (ICICC) 2021 (April 22, 2021).
  • 🔗 [Link to Publication - You'll need to find and add the actual URL or DOI link]
• Mental Workload Estimation Using EEG:
  • Published in: 2020 Fifth International Conference on Research in Computational Intelligence and Communication Networks (ICRCICN), Bangalore, India, 2020.
  • 🔗 [Link to Publication - You'll need to find and add the actual URL or DOI link, e.g., IEEE Xplore link]

I'm always open to discussing new projects, research ideas, or collaboration opportunities.