Django: profile memory usage with Memray

Memray is a memory profiler for Python that helps track memory allocation and deallocation through flame graphs. To profile a Django project, start by running `memray run manage.py check` to measure startup memory usage, then generate a flame graph with `memray flamegraph`. The visualization shows memory allocations over time, with frame width proportional to memory used. Common optimizations include deferring expensive imports like numpy until needed, using lazy imports, or replacing heavy dependencies with lighter alternatives. A practical example demonstrates reducing startup memory from 25.2 MB to 19.4 MB by replacing numpy's shuffle with Python's built-in random.shuffle().