AI Research | Machine Learning | NLP | Multimodal Systems

I am a Mechanical Engineering graduate from IIT Kanpur with a strong research orientation at the intersection of machine learning, natural language processing, and multimodal AI systems. My work focuses on translating rigorous engineering foundations into modern, research-grade AI solutions.

I am particularly interested in representation learning, reasoning with large language models, and applied ML systems that operate under real-world constraints such as noisy data, limited supervision, and domain specificity.

• Large Language Models (LLMs) and fine-tuning techniques (QLoRA, PEFT)
• Retrieval-Augmented Generation (RAG) systems
• Multimodal learning (text–image embeddings)
• Agentic and reasoning-oriented AI systems
• Applied NLP for enterprise and document intelligence
• Robust ML under noisy and imperfect data (OCR, weak labels)

• LLM Fine-Tuning Fine-tuned large language models using QLoRA for domain-specific tasks, focusing on parameter efficiency and deployment feasibility.

• RAG Pipelines (Text + Image) Designed and implemented retrieval-augmented generation pipelines leveraging text-image embeddings for knowledge-intensive tasks.

• Agentic Systems Built agent-based workflows for reasoning and task decomposition using modern orchestration frameworks.

• Document Intelligence Applications

  • Invoice field prediction under noisy OCR conditions
  • Purchase Order (PO) matching and reconciliation
  • Domain-specific AI assistants for enterprise use cases

Tools & Frameworks: LangChain · FAISS · Hugging Face · CLIP · Transformers · PyTorch

My academic background includes strong exposure to:

• Robotics
• Signal Processing
• Control Systems
• Classical Mechanical Engineering analysis

This foundation enables me to approach AI problems with system-level thinking, mathematical rigor, and experimental discipline.

• Languages: Python, MATLAB, Dart (Flutter)
• ML/DL: PyTorch, Hugging Face, PEFT, Transformers
• Data & Systems: FAISS, Vector Databases, OCR Pipelines
• App Development: Flutter (e.g., weight conversion app project)

I thrive in research-driven environments where I can:

• Prototype and iterate rapidly
• Explore new methodologies and architectures
• Work on applied ML/NLP problems with real-world impact

I am open to:

• Research collaborations
• Applied AI / ML projects
• Open-source research-oriented work

If you are working on applied ML, NLP, multimodal systems, or research-focused AI projects, feel free to reach out. I am always interested in meaningful collaborations and intellectually challenging problems.

“Bridging classical engineering rigor with modern AI experimentation.”