Bhuwan
Oct 2025 - Present
MSc in Natural Language Processing
- Building foundational knowledge in computational linguistics, machine/deep learning, transformers, LLMs, and GenAI.
- Developing projects using Python and NLP libraries such as Hugging Face Transformers, spaCy, PyTorch, and NLTK for text processing and model building.
- Researching code-generation principles in NLP, an intersection of AI and software engineering; wrote a term paper on how autoregressive language models perform Fill-In-the-Middle (FIM) code generation.
- Researched test-time compute scaling in small reasoning LLMs (Qwen3) with vLLM, comparing long chain-of-thought against parallel sampling with majority voting across 13K+ reasoning traces.
- Building LLM applications with Retrieval-Augmented Generation (RAG), embeddings, semantic search, and vector databases, plus agentic AI workflows using LangChain, tool calling, and MCP.
- Exploring LLM fine-tuning (LoRA/PEFT), LLM evaluation, prompt engineering, MLOps, and Graph databases.
