Next
Memory context snapshots
Record per-request memory exposure for prompt builds, recall tools, smart reads, memorize internals, and dream runs so teams can see which memory actually influenced an agent or model call.
Public roadmap
Tell us what memory workflows matter for your AI product.
This roadmap is directional, not a promise board. Register interest, leave a note about your use case, or schedule time with the founder if your product has a real memory problem today.
Now / shipped
The foundation is already live.
Postgres-backed memory files
Revision history and access logs
Files time travel in the admin UI
Admin UI and Playground
TypeScript and Python SDKs
MCP over SSE and stdio
Scoped user/ and shared/ memory
Opt-in shared writable memory
Hybrid pgvector search with reciprocal rank fusion
Next
Near-term work that makes memory more trustworthy in production.
These are the features most closely tied to operating durable user memory once real usage starts accumulating.
Next
Memory health signals
Surface review-worthy memory such as files never used in 30 days, written by one-off agents, missing owners, frequently injected but never updated, or possibly contradicted by newer memory.
Next
Ownership and TTL controls
Give memory files owners, review status, and expiry windows so stale or ownerless memory becomes a visible governance workflow instead of silent drift.
Next
PII hooks
Redact, block, or review sensitive memory before writes land in service mode or direct Postgres mode.
Next
Team memory proposals
Add a review queue on top of shared writable mode so agents can propose shared-memory updates while admins review, accept, reject, or auto-approve contributions.
Next
Post-write hooks
Trigger webhooks after memory changes so Slack, n8n, Zapier, app events, and audit stores can stay in sync.
Next
Bidirectional memory links
Index memory links as a precomputed backlink graph. With reverse traversal, seeding on a hub file can surface the most recent notes that reference it, and highly referenced hub files can rank higher in search.
Exploring
Bigger bets we will only build if the pull is real.
Some teams need richer retrieval, broader isolation boundaries, or background synthesis. Tell us which of these would change your roadmap.
Exploring
Sidecar memory writes
Pass a raw_data argument to memory_write so bulk transcripts, documents, or structured payloads can be written directly to Postgres without routing the full payload through the model context window.
Exploring
Per-user dreaming budgets
Track daily background consolidation spend per user and skip dream cycles once the configured budget is exhausted.
Exploring
Per-file dreaming exclusions
Let operators mark sensitive or hand-curated memory files as excluded from background dreaming without relying only on global path rules.
Exploring
Named mounts — team, org, workspace scopes
Register additional memory scopes at init time. A team agent writes to team/itinerary.md and user/prefs.md simultaneously — both isolated, both in the same call. No migration needed.
Exploring
Confidence and provenance on memories
Attach a confidence score (0–1) and source type (explicit, inferred, consolidated) to each memory file. Explicitly stated facts outrank agent-inferred ones. The dream consolidation job propagates confidence through merges and can downgrade contradicted memories rather than deleting them. Ranking in search and smart_read multiplies by confidence — stale or uncertain memories fade, high-signal ones stay front-of-mind.
Exploring
Source-scoped memory
Track where facts came from and organize durable memory by app events, documents, projects, and conversations.
Done / absorbed
Recent roadmap work has moved into shipped memory infrastructure.
Files time travel is now part of the admin workflow. Background dreaming also handles the cleanup work that used to sit behind separate memory health and compaction roadmap items. We may still add sharper diagnostics later, but the default product direction is to make memory maintenance inspectable and automatic.
Files time travel
Operators can open the Files tab with an As of timestamp, browse the historical file tree reconstructed from revisions, and diff matched historical content against the current file.
Dreaming and background synthesis
Shipped as opt-in service-mode consolidation. Dreaming waits for quiet memory, merges duplicates, clarifies fragmented notes, resolves direct contradictions, and writes normal revisions as dream-agent.
Admin Dreams panel
Operators can read dream config, update cadence and write budgets, inspect per-user dream status, and pause or resume dreaming for specific users.
Dream-safe audit trail
Dream runs skip log files, avoid no-op dream-log entries, record files_touched, and preserve the same revision/access-log trail as normal memory writes.
De-prioritized
See how dreaming covers thisAutomatic-only memory health
De-prioritized as a standalone cleanup-only feature because dreaming covers part of the background maintenance loop. Diagnostics, audit snapshots, health signals, and operator review are now reprioritized as production trust features.
De-prioritized
See how dreaming covers thisMemory compaction
De-prioritized as a separate feature because dreaming already summarizes, deduplicates, and keeps long-running memory files readable while preserving revisions as the original trail.
Founder call
Building an AI product where users feel forgotten?
A short founder call is the fastest way to map your current memory approach, the retention pain, and whether MemexAI should fit now or later.
Roadmap votes are product signal, not commitments. We use them to prioritize conversations and choose what to make legible, durable, and self-hostable next.