My research foundation is built on a strong understanding of seminal research papers, specifically the "Attention Is All You Need" paper.
I possess advanced knowledge of Multi-Head Attention mechanisms and modern optimization techniques including MQA (Multi-Query Attention), GQA (Grouped-Query Attention), MLA (Multi-Head Latent Attention), and FlashAttention .
I specialize in comprehensive LLM Architecture design, covering the entire pipeline from pre-training and fine-tuning and deployment. My work extends to building sophisticated Generative AI and Agentic AI systems that solve complex problems with efficiency and precision.
Built a Transformer-based Large Language Model including tokenizer, embeddings, and attention blocks.
Implementation of the self-attention layer from "Attention is All You Need" paper.
Retrieval Augmented Generation system for medical documents using FAISS & Groq.