LESSON 7.3 · Systems · 120 min
Context engineering and memory
Context engineering and memory. Learn working context, retrieval, summaries, and memory policies and complete: Compare three memory strategies. Part of the “From Models to Intellig
Learning objectives
- Explain what problem “Context engineering and memory” solves without hiding behind terminology.
- Trace the variables and causal links across working context, retrieval, summaries.
- Complete “Compare three memory strategies” and judge the result with evidence rather than intuition.
Core concepts
working context
working context is part of the lesson’s causal model. State its inputs, outputs, invariants, and failure mode; then verify it with a hand-check or a minimal experiment before moving to an optimized implementation.
retrieval
retrieval is part of the lesson’s causal model. State its inputs, outputs, invariants, and failure mode; then verify it with a hand-check or a minimal experiment.
summaries
summaries is part of the lesson’s causal model. State its inputs, outputs, invariants, and failure mode; then verify it with a hand-check or a minimal experiment.
and memory policies
and memory policies is part of the lesson’s causal model. State its inputs, outputs, invariants, and failure mode; then verify it with a hand-check or a minimal experiment.
Build and verify
Compare three memory strategies
- Predict: write the expected output, trend, or failure before running code.
- Build: implement only the minimum components needed to answer the question.
- Verify: compare with a baseline or trusted implementation; save seeds, parameters, and raw outputs.
- Transfer: change one shape, dataset, scale, or workload condition and explain whether the conclusion still holds.