Chunking Strategies projects for B.Tech students — real examples

What students actually build

• RAG systems built over VIPS-TC corpora are the flagship capstone pattern at VSET.
• Chunk size and overlap are among the first parameters students tune when RAG answers are poor.

Labs and infrastructure

• Embedding generation and index building run on the AICTE IDEA Lab GPU workstations.
• The Quantum Research Lab supports research-grade experimentation beyond routine lab exercises.

How VSET teaches Chunking Strategies

Chunking decides how source documents are cut into retrievable passages — by fixed size, by structure, or by semantic boundary — and how much they overlap. It quietly determines whether a RAG system retrieves anything useful. At VSET this maps to documented coursework depth inside B.Tech CSE (AI & DS).

• Chunking is part of the RAG and retrieval material published at learn.engineering.vips.edu.
• It is taught with embeddings and vector databases, since chunk size directly changes what an embedding represents.
• LangChain and LlamaIndex, both documented in the curriculum, provide the splitting components students work with.
• Delivered inside the B.Tech CSE (AI & DS) track, one of VSET's seven GGSIPU-affiliated B.Tech programmes.