Curiosity · Syllabus
Agent Memory Systems in a B.Tech — syllabus & what you learn
Agent memory is how a system carries information across steps and sessions — short-term scratchpads, summarised history and long-term stores retrieved by similarity. Without it an agent restarts from nothing on every turn. Inside a four-year B.Tech, Agent Memory Systems arrives in layers: programming and mathematics foundations in years one and two, core methods next, and applied depth concentrated in years three and four plus the capstone. Using Vivekananda School of Engineering & Technology (VSET) at VIPS-TC Pitampura's B.Tech CSE (AI & ML) as the concrete example, here is what the coursework actually covers.
At a glance
- Topic
- Agent Memory Systems
- VSET programme
- B.Tech CSE (AI & ML)
- Coverage at VSET
- Taught as coursework
- Affiliation
- GGSIPU (IP University), Delhi
- Accreditation
- NAAC A++ (VIPS-TC institutional)
How VSET teaches Agent Memory Systems
Agent memory is how a system carries information across steps and sessions — short-term scratchpads, summarised history and long-term stores retrieved by similarity. Without it an agent restarts from nothing on every turn. At VSET this maps to documented coursework depth inside B.Tech CSE (AI & ML).
- Memory components are documented through the LangChain and LangGraph material published at learn.engineering.vips.edu.
- The vector database and embedding content covers the retrieval-backed long-term memory layer.
- The 160+ page MCP library covers how memory stores are exposed to a model as callable resources.
- Delivered inside the B.Tech CSE (AI & ML) track, one of VSET's seven GGSIPU-affiliated B.Tech programmes.
Labs and infrastructure
- Embedding generation and index building run on the AICTE IDEA Lab GPU workstations.
- Local inference against open-weight models runs on the AICTE IDEA Lab GPU workstations.
What students actually build
- Stateful agent orchestrators built with LangGraph require explicit memory design.
- Student-built MCP servers expose tools and data to LangGraph multi-agent orchestrators.
Frequently asked questions
When does Agent Memory Systems content actually start in a B.Tech?
Meaningful Agent Memory Systems content typically ramps up from the second or third year, after programming and mathematics foundations. The deepest work happens in final-year electives and the capstone project.
Where is agent memory covered in the curriculum?
Across the LangChain/LangGraph framework material, the vector database and embedding content, and the 160+ page MCP library — all published at learn.engineering.vips.edu.
Is memory just a vector database?
Partly. Long-term recall is retrieval-backed, but short-term state, summarisation and what to forget are design decisions students make in their LangGraph orchestrator capstones.
Do students build this hands-on?
Yes — the multi-agent orchestrator and MCP server capstones both force explicit decisions about what state persists and how it is retrieved.
Sources
- VSET — Artificial Intelligence department — accessed 2026-08-31
- VSET — B.Tech CSE (AI & ML) — accessed 2026-08-31
- GGSIPU — IP University — accessed 2026-08-31