Decomposition rules for bounded-context scheduling
Type: note · Status: seedling
These rules follow from the symbolic scheduling model. They are preliminary — we expect to discover more as the model develops.
What is being optimised
The optimisation problem is:
Choose a decomposition, a prompt-construction strategy, and a schedule of state transformations that maximises expected task utility while respecting the per-call bound M.
Even before underspecified semantics enter, there are several objective terms:
- total token traffic between scheduler state and LLM calls
- number of agent calls
- peak prompt size
- information loss from compression
- preservation of cross-item interactions needed by later synthesis
This makes the problem different from ordinary knapsack-style context packing. The scheduler must trade off:
- early filtering against the risk of discarding something that matters later
- aggressive summarisation against the risk of destroying interactions needed for synthesis
- many narrow calls against the overhead of orchestration
- loading raw state-derived material against saving task-shaped intermediate ones
The first two are about optionality — paying context now to keep options open later. The latter two are about cost structure — choosing between representations and decompositions with different efficiency profiles. Both kinds of trade-off are present in every scheduling decision.
Rules
Separate selection from joint reasoning. First use cheap narrow calls to discover sparsity. Only then pay for wide calls that need multiple items together.
Use symbolic operations wherever exactness is available. Retrieval, thresholding, sorting, prompt assembly, and name-based routing should be outside the LLM window whenever possible.
Save reusable intermediate items in scheduler state. Relevance labels, extracted claims, and task-specific summaries are worth keeping when they are much cheaper to reuse than reconstructing the originals.
Delay expensive co-loading until interactions justify it. Joint loading is valuable only when the task depends on relations between items rather than independent judgments about them.
Commit low-degree-of-freedom choices first. When one decision has only a narrow feasible set and another has many workable options, decide the constrained one first. This preserves optionality and avoids consuming scarce valid placements too early.
Do not compress away needed interfaces. If the final answer depends on tensions, contradictions, or alignments between sources, summaries should preserve pointers or extracted structures that keep those interactions recoverable.
Choose representations, not just subsets. The main optimisation variable is often not which notes to load, but whether to expose bodies, extracts, summaries, or previous synthesis items to the bounded call.
Exploit clean frames recursively. When the relevant set is still too large, apply the same pattern again: filter, cluster, compress, and merge in a tree rather than a flat history.
Open Questions
- Which classes of lossy derived items preserve enough structure for later synthesis?
Relevant Notes:
- symbolic scheduling over bounded LLM calls is the right model for agent orchestration — foundation: the model these rules follow from
- context efficiency is the central design concern in agent systems — cost model: context is the scarce resource these rules optimise over
- distillation — mechanism: saved intermediate items are often distillations shaped for later reuse
- solve low-degree-of-freedom subproblems first to avoid blocking better designs — extends: general ordering heuristic that explains why constraint-setting should happen before flexible synthesis choices
Topics: