Skip to content

Memory Retrieval Architecture v2 (Epic) #21

@aurexav

Description

@aurexav

Context
We are redesigning the memory retrieval architecture to match the strongest known patterns from qmd and claude-mem, while keeping ELF invariants (evidence binding, deterministic writes, and rebuildable indexing).

Goal
Deliver a best-in-class retrieval pipeline focused on recall, precision, ranking stability, and controllable context growth.

Scope (tracked by sub-issues)

Backlog (explicitly out of scope for this epic)

Acceptance criteria

  • All in-scope sub-issues are delivered.
  • Each change that affects ranking is evaluated with the retrieval harness (same dataset, before/after comparison).

References

Metadata

Metadata

Assignees

No one assigned

    Labels

    area:serviceRetrieval logic, ranking, and request orchestration.area:storagePostgres schema, SQL queries, and storage correctness.area:workerBackground workers, outbox processing, and indexing pipelines.kind:epicUmbrella issue that tracks multiple deliverables.theme:evaluationQuality measurement, gold sets, regressions, and metrics.

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions