Each unique task title is now enriched by LiteLLM once and cached in the DB.
Subsequent agent compute cycles (every 12h) fetch the cache before calling
ml-serving; only new titles hit the tip-generator.
- DB: task_enrichments(content_hash PK, description, model, created_at)
- TS: fetchEnrichmentCache / persistEnrichments helpers in agent-outputs.ts;
enrichment_cache passed in compute request, new_enrichments persisted from response
- Python: AgentComputeRequest.enrichment_cache / AgentComputeResponse.new_enrichments;
AgentInput.enrichment_cache; _enrich_batch returns (descriptions, new_entries);
cluster_tasks returns (clusters, new_enrichments)
- FocusAreaAgent stashes new_enrichments in signals_snapshot under _new_enrichments;
compute_agent endpoint pops it before storing the snapshot
Closes part of #129
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
When fetchOrchestratorTip returned null (LiteLLM timeout, bad JSON, etc.)
the recommender silently fell back to randomPolicy, serving a raw Todoist
task with no rationale — explaining both reported symptoms.
- Remove randomPolicy/signalToCandidate; return 204 when orchestrator fails
so the UI shows "All clear" instead of a confusing Todoist task
- Pass recent_tip through the stack (frontend → POST /recommend →
fetchOrchestratorTip → ml/serving RecommendRequest → build_orchestrator_messages)
so after snooze the LLM is instructed not to repeat the snoozed content
Fixes#122
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Convert ml-serving from isolated MLflow runs to nested traces using
mlflow.start_span_no_context(). The recommend endpoint now emits a full
span tree: recommend (CHAIN) → build_context (TOOL), agent:* (AGENT) ×N,
llm_orchestrator (LLM). Compute and infer endpoints each emit a single span.
Supporting changes:
- mlflow-skinny>=3.1.0 added to requirements
- MLflow configured with --serve-artifacts + mlflow-artifacts:/ default root
for cross-container artifact proxy (spans now persist from ml-serving)
- --allowed-hosts extended to include mlflow:5000 (SDK includes port in Host)
- science_destiny slider wired through prompts.py and recommend endpoint
- Config page exposes science/destiny slider (0=data-driven, 100=intuitive)
- Tip page shows rationale inline on tap
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Pass MLFLOW_ADMIN_PASSWORD as fallback password credential
- Set host_header='localhost' to satisfy MLflow's --allowed-hosts check
(MLflow rejects Host: mlflow but accepts Host: localhost)
- Default MLFLOW_TRACKING_URI to http://mlflow:5000 in compose so the
env_file value is not silently overridden to empty
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Logs one MLflow run per /recommend (params, token metrics, latency,
full prompt + tip as artifacts) and per /agents/{id}/compute and
/infer call (signals snapshot, inferred prefs, latency).
Tracing is a no-op when MLFLOW_TRACKING_URI is unset; ml-serving
starts and serves tips correctly without MLflow configured.
Refs #118 (M4: remove from production / move off critical path).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Replaces snooze-rate heuristic with p50 of actual task lateness (completedAt − dueAt).
Adds project_realness inference: projects with chronic lateness get realness < 1 and
the agent softens its snippet language from "overdue" to "past target date".
- TaskCompletion added to UserHistory with lateness_days computed property
- _infer_lateness_tolerance: p50 of task_completions, clipped at 0, float
- _infer_project_realness: per-project median lateness normalised by global median
- Both InferredParams use 7d TTL; cold_start = 0.0 / {}
- AgentInferRequest accepts task_completions; endpoint wires them through
- 12 new tests covering punctual/chronic/mixed users and language softening
- Agent bumped to v1.2.0
Closes#115
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Each agent now exports a module-level MANIFEST declaring id, version,
pref_schema, required_consents, ttl_sec, and silenced_in_contexts. The
registry surfaces both the agent and its manifest, and rejects on
mismatch so the two cannot drift.
ml/serving exposes GET /agents/registry; services/api proxies it as
GET /api/agents/registry with a 60s in-process cache so admin pageviews
don't hammer upstream. Failures aren't cached.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Drop all four Airflow containers (db, init, webserver, scheduler) from the
mlops compose profile, leaving MLflow as the sole mlops service. Remove
AIRFLOW_* env vars, config fields, health-check entries, DAG trigger code
in admin/bench routes, the airflow_dag_run_id schema column, Airflow nav
links and DAG-run links in the admin UI, the two Airflow DAG files
(bench_dag.py, sim_dag.py), and all related docs/ADR references.
Simulations now run exclusively via the subprocess path.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- prompts.py: sort tasks overdue-first → priority desc → age desc before
rendering into the LLM prompt (same ordering as ml/features/context.py)
- prompts.py: render User profile summary line (completion_rate, dismiss_rate,
preferred_hour) when profile_features are present
- main.py: add profile_features field to PromptContext; plumb from
GenerateRequest into the prompt builder via model_copy
- logging_config.py: drop add_logger_name processor (incompatible with
PrintLoggerFactory — caused test ordering failures)
- test_generate.py: 6 new tests covering sort order, profile rendering,
partial fields, empty profile, and end-to-end plumbing through /generate
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- TS: pino + pino-http; every HTTP request log includes traceId from
W3C traceparent header (generated if absent); forwarded to ml/serving
on all /score, /generate, /reward, and /api/ml proxy calls
- Python: structlog JSON; FastAPI middleware binds trace_id via
contextvars so every log line within a request carries it
- Sentry: optional SENTRY_DSN init in both runtimes (no-op if unset)
- Replace all console.* calls across services/api with pino logger
- Update tests to spy on logger instead of console
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Adds a NATS JetStream consumer to ml/serving so the feature pipeline
can react to events without the API triggering every read.
- nats_consumer.py: durable push consumers for signals.> and feedback.>
streams; acks on success, naks for redeliver, up to NATS_MAX_DELIVER
attempts; per-consumer health state (last_msg_ts, processed, errors)
- main.py: FastAPI lifespan wires start/stop; /health exposes nats state
- requirements.txt: adds nats-py>=2.9.0
- Dockerfile.ml: copy all *.py from ml/serving (was missing prompts.py)
Handled subjects:
signals.task.synced → writes per-user sync metadata to STATE_DIR
signals.tip.feedback → logged for observability (reward via HTTP path)
Config: NATS_URL (empty = disabled), NATS_DURABLE_PREFIX, NATS_MAX_DELIVER
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Ship the scaffolding for #99 (phase B.3 of #81):
- ml/serving: add /score/egreedy/v2, /reward/egreedy/v2, /stats/egreedy/v2
endpoints (D=12). New feature dims: completion/dismiss rates, mean dwell
(clipped 10min), preferred-hour alignment (cosine, 1-dim), tip volume (log).
Separate state file per user (_egreedy_v2.json). /reset clears v2 state too.
- ADR-0012: documents D=7→12 dimension change, normalization choices, shadow
rollout protocol, and promotion gate (offline sim win per ADR-0002).
- recommender.ts: register egreedy-v2-shadow in shadow-policy map (disabled by
default). When enabled, calls /score/egreedy/v2 fire-and-forget and publishes
shadow:egreedy-v2-shadow serve signal. No reward to shadow — sim is the gate.
- sim runner/personas: personas carry synthetic profile_features per persona;
_call_score/_call_reward thread profile_features through (None-safe for v1/linucb).
- 18 new Python tests; all 56 Python + 170 TS tests pass.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Centralizes user-level features (completion_rate_30d, dismiss_rate_30d,
mean_dwell_ms_30d, preferred_hour, tip_volume_30d) in a TS registry that
owns both definition and SQL aggregation, since the data lives in the
TS-owned SQLite tables (tip_views/tip_feedback). Lazy TTL refresh keeps
recommend latency bounded; values persist in user_profile_features (KV).
ml/serving accepts profile_features on /score + /generate but does not
yet consume them — extending the bandit feature vector changes D and
resets every user's learned state, so that's a deliberate phase-B step.
Includes ml/features/profile_schema.py as a contract mirror with a sync
test that diffs name sets against registry.ts.
ADR-0011 records the data-locality reasoning (registry in TS, not Python
as the issue originally suggested).
Phase B (deferred): event-driven incremental updates, bandit consumption
with state migration, admin per-user profile page, staleness alerts.
Refs #81.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Replaces the hardcoded "v1" label with a real prompt registry:
ml/serving/prompts.py — keyed by version: v1 (baseline),
v2-mentor (calm/specific persona),
v3-few-shot (v1 persona + curated examples)
ml/serving/main.py — POST /generate accepts optional prompt_version,
422 on unknown, echoes the version actually used
back in the response
services/api/src/config.ts — TIP_PROMPT_VERSION: empty / single / comma-list
(uniform random per request)
services/api/src/routes/recommender.ts
— pickPromptVersion() drives selection; the
response's prompt_version (not a stale TS
constant) is what lands in tip_scores so the
#92 reward-analytics dashboard shows real
per-variant reaction rates
Closes#84.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
ML serving:
- LinUCB contextual bandit (disjoint, d=5 features: hour_sin/cos, is_overdue, task_age, priority)
- /score endpoint replaces stub random; /reward endpoint for online learning
- Per-user model state persisted to disk as JSON (survives restarts)
- venv at ml/serving/.venv; start with pnpm dev from ml/serving
Recommender:
- Todoist fetch now extracts features (is_overdue, task_age_days, priority)
- RemotePolicy calls ml/serving with 3s timeout; falls back to RandomPolicy
- Reward sent to /reward on feedback (done=+1, snooze=0, dismiss=-1)
Web Push:
- VAPID keys in config; push_subscriptions table in DB
- POST/DELETE /api/push/subscribe; GET /api/push/vapid-public-key
- Service worker (public/sw.js): push → showNotification, notificationclick → focus/open
- "notify me" button on tip page; registers SW + subscribes on permission grant
Event bus:
- services/api/src/events/bus.ts: typed EventEmitter wrapper
- Subjects: signals.tip.served, signals.tip.feedback, signals.task.synced
- Same publish/subscribe API NATS JetStream will implement — swap is mechanical
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
