132 Commits

Include renames

Author	SHA1	Message	Date
Celes Renata	af7b654fb7	fix: vLLM deployment — Recreate strategy, correct service port, single model name ... - Strategy: Recreate (GPU can only be used by one pod at a time) - nuextract-external targetPort: 8080 (nginx proxy port) - served-model-name: numind/NuExtract3 - DB ai_agents updated to model=numind/NuExtract3 provider=vllm	2026-07-04 05:34:04 +00:00
Celes Renata	a322e00659	feat: vLLM serving numind/NuExtract3 on full GPU, ollama disabled ... - Ollama scaled to 0 — vLLM gets entire 16GB GPU - NuExtract3 (9B) running at 80% GPU utilization - Both OLLAMA_BASE_URL and VLLM_BASE_URL point to vLLM service - Fixed gremlin-1 missing default route (no internet access) - Removed GPU resource limit (time-slicing handles allocation) - Removed init container (direct download works with internet fix)	2026-07-03 17:57:19 +00:00
Celes Renata	ecade0dd52	feat: vLLM in K8s via Harbor mirror, pipelines point to internal svc ... - vLLM image mirrored to registry.celestium.life/stonks-oracle/vllm-openai - Deployment uses Harbor image (Docker Hub IPv6 unreachable from cluster) - All 3 pipelines use vllm-external.vllm-service.svc.cluster.local:2701 - K8s manifests at infra/kube-vllm/ and synced to ~/sources/kube/vllm	2026-07-03 17:02:31 +00:00
Celes Renata	3a9894cd03	fix: revert VLLM to bare-metal 192.168.42.254:2701 ... K8s vLLM deployment blocked by Docker Hub IPv6 connectivity issue. Falling back to NixOS systemd vLLM service on esnixi. K8s manifests remain in infra/kube-vllm/ for future use.	2026-07-03 11:40:36 +00:00
Celes Renata	4ac10ab4f6	fix: point VLLM_BASE_URL to K8s internal vllm-service ... vLLM now runs in vllm-service namespace with metrics proxy, accessible at vllm-external.vllm-service.svc.cluster.local:2701	2026-07-03 11:38:24 +00:00
Celes Renata	3996ec6ef5	feat: switch models — ollama qwen3.5:4b-fast, vllm numind/NuExtract3 ... - Ollama: qwen3.5:4b-fast with NUM_PARALLEL=8, num_ctx=16384 (~6GB VRAM) - vLLM: numind/NuExtract3 at 192.168.42.254:2701, gpu-memory-util=0.45 - All 3 pipelines (beta/paper/live) updated with correct endpoints - Ollama proxy at 10.1.1.12:2701 (K8s ollama-metrics) - vLLM proxy at 192.168.42.254:2701 (NixOS nginx)	2026-07-03 11:15:28 +00:00
Celes Renata	14a9b4fcc1	fix: ops pipeline — rate limit, validation cycle, snapshots, config query, rejection reason ... - Polygon rate limit env-configurable, default 5 (free tier) - v3_engine_enabled reads from JSONB config column correctly - Validation cycle (outcome eval + metrics) wired into scheduler hourly - Daily portfolio/risk snapshots after 16:30 ET with idempotency - Prediction snapshot price fallback from positions table - Order rejection_reason + rejected_at persisted on insert - Lake-publisher scaled to 0 replicas (redundant) - Test fix: isolate per-type rate limit tests from Polygon global	2026-07-03 08:29:24 +00:00
Celes Renata	b70304ad6c	fix: correct Alpaca paper trading credentials for beta	2026-06-30 15:24:30 +00:00
Celes Renata	fdb54b77ce	fix: update Redis password, Ollama URL, and broker keys for new cluster	2026-06-28 22:03:26 +00:00
Celes Hillyerd	bf9b7e65dd	revert: restore registry cache, investigate LE cert	2026-06-26 03:28:59 -07:00
Celes Hillyerd	a0eaa7d6c8	ci: pull from Docker Hub directly, remove broken registry cache/mirrors	2026-06-26 03:27:12 -07:00
Celes Renata	18f857e1a3	feat: add live-math stage with dedicated paper trading account ... New ArgoCD stage 'stonks-live-math' running the dual-pipeline signal engine against a separate Alpaca paper account for production validation. - Namespace: stonks-live-math - Database: stonks_live_math (Redis DB 3) - Broker: PK64RS7NH24XPBI3IDEU3BB72Y (paper-api.alpaca.markets) - Ingress: stonks-math-*.celestium.life - Deploys alongside existing live stage (same image tag)	2026-05-02 08:19:47 +00:00
Celes Renata	f468e30af0	feat: implement dual-pipeline signal engine service ... New service at services/signal_engine/ implementing concurrent heuristic (deterministic scoring) and probabilistic (Bayesian inference) pipelines that evaluate technical signals across 6 timeframes (M30-M) and produce independent BUY/WATCH/SKIP verdicts per ticker per evaluation tick. Components: - Input Normalizer: multi-source data assembly with sentinel fallbacks - Signal Library: Fibonacci, MA Stack, RSI, Cup & Handle, Elliott Wave - Multi-Timeframe Confluence Engine: weighted scoring with D/W/M anchors - Hard Filter Engine: macro_bias, valuation, earnings proximity gating - Heuristic Pipeline: S_total scoring with confidence-gated verdicts - Probabilistic Pipeline: Bayesian log-odds with regime priors, entropy gating, EV_R calculation, and signal correlation penalty - Exit Engine: stop-loss, targets, trailing ATR-based stops - Delta Analyzer: pipeline agreement tracking with rolling Redis metrics - Output Formatter: SignalOutput contract + Recommendation schema mapping - Worker orchestrator: concurrent pipelines with failure isolation - Main entry point: queue polling with fail-safe config loading Infrastructure: - Migration 039: signal_engine_outputs table with 3 indexes - Helm chart: signalEngine service entry (processing tier) - Redis key: QUEUE_SIGNAL_ENGINE constant Tests: 390 tests (unit + property-based) covering all components Config: dual_pipeline_enabled=false by default (safe rollout)	2026-05-02 07:32:26 +00:00
Celes Renata	bc077bfcc8	feat: trading feedback engine — periodic performance reports with AI summarization ... - Migration 038: trading_reports table + report-summarizer agent seed - 6 reporting modules: models, collector, sections, validator, summarizer, generator - API endpoints: GET /api/reports (paginated, filterable), GET /api/reports/{id} - Frontend hooks: useReports, useReport with TanStack Query - Scheduler: daily (after 16:30 ET) and weekly (Saturday) report triggers - Redis queue consumer for async report generation with retry/dedup - 5 property-based tests (chunking, serialization, validation, accuracy, deltas) - 109 unit/integration tests across all modules - 6 frontend hook tests with MSW mocks	2026-05-01 22:13:09 +00:00
Celes Renata	376fcb4bb4	feat: add 12 integrity check saved queries for SQL Explorer (migration 037) ... Prefixed with ⚕ for easy identification. Checks: duplicate snapshots, orphaned links, evidence count mismatches, contribution score sums, canonical key consistency, out-of-range values, unmatched snapshots, zero evidence rate, duplicate count mismatches, missing prices, outcome integrity, and pipeline health summary.	2026-05-01 19:20:56 +00:00
Celes Renata	cc21fd9e8f	feat: add 15 validation saved queries for SQL Explorer (migration 036) ... Prediction overview, confidence distribution, evidence dedup quality, source/catalyst breakdown, win rate by ticker/horizon, model quality timeline, quality gate status, high-duplicate predictions, excess vs SPY.	2026-05-01 19:15:54 +00:00
Celes Renata	affb65d7f4	fix: point VLLM_BASE_URL to Ollama (10.1.1.12:2701) and update model name ... vLLM service on 192.168.42.254:8000 was down causing extraction failures. Also updated ai_agents table to use ollama provider directly.	2026-05-01 18:56:55 +00:00
Celes Renata	0732894414	fix: widen horizon/mode columns in prediction_snapshots to VARCHAR(50) ... time_horizon values like 'position_30d_90d (proj:30d)' exceed VARCHAR(20)	2026-05-01 04:27:19 +00:00
Celes Renata	76f6bd5677	fix: quote reserved keyword 'window' in prediction_snapshots SQL	2026-05-01 03:40:48 +00:00
Celes Renata	7fcc8a6c07	feat: model validation, calibration, and signal quality layer ... - Migration 035: prediction_snapshots, prediction_outcomes, signal_evidence_links, model_metric_snapshots tables + SQL views - Prediction snapshot writer with canonical evidence keys, duplicate detection, contribution scores - Outcome evaluator across 5 horizons (1h, 6h, 1d, 7d, 30d) - Metrics engine: ECE, Brier score, IC, Rank IC, benchmark comparison - Attribution engine: per-source, per-catalyst, per-layer performance - Calibration engine: Bayesian shrinkage source reliability - Quality gate for live trading eligibility with configurable thresholds - 7 new /api/validation/* endpoints - Upgraded OpsModel dashboard with validation tab - Enhanced recommendation display with calibration context - Backtest replay validation mode - 86 Python tests (unit + property-based), 179 frontend tests passing	2026-05-01 03:04:58 +00:00
Celes Renata	6169efdc89	fix: copy scripts/ into Docker image + graceful backfill fallback ... - Dockerfile now copies scripts/ directory into /app/scripts/ - Init container uses sh -c with fallback so missing script doesn't crash the pod (graceful degradation until image rebuilds)	2026-04-30 23:18:43 +00:00
Celes Renata	c42f2223d8	fix: move backfill logic to standalone script (fixes YAML parse error) ... Inline Python with f-strings and colons broke Helm YAML parsing. Moved to scripts/backfill_market_data.py and call it directly from the init container command.	2026-04-30 23:01:20 +00:00
Celes Renata	414f476620	fix: auto-backfill market data on startup + fix trading dedup bug ... 1. Add backfill-market-data init container to scheduler deployment. On vanilla start (< 50 bars in market_snapshots), fetches 90 days of daily OHLCV from Polygon directly via asyncpg + httpx. 2. Fix trading engine dedup key placement. Previously the Redis dedup key was set BEFORE evaluate_recommendation(), so recs skipped as outside_trading_window were permanently deduped and never retried when the market opened. Now the dedup key is only set AFTER the decision for non-retryable outcomes (act, confidence_too_low, etc). outside_trading_window skips are retryable.	2026-04-30 14:08:54 +00:00
Celes Renata	a5f2bcde55	fix: set MARKET_DATA_BASE_URL to polygon in base values.yaml	2026-04-30 07:20:13 +00:00
Celes Renata	601b85764b	fix: mount stonks-market-secrets on query-api for backfill endpoint	2026-04-30 06:33:26 +00:00
Celes Renata	fa18b1a7c2	fix: populate broker and market API secrets in base values.yaml ... ArgoCD was syncing empty secrets back over manual patches. Credentials now match stonks-paper (same Alpaca paper + Polygon keys).	2026-04-30 05:05:43 +00:00
Celes Renata	4b254adad2	fix: scale ingestion and extractor to 1 replica in prod	2026-04-30 05:04:20 +00:00
Celes Renata	af2e554edd	fix: switch ai_agents seed from vllm to ollama qwen3.5:9b-fast ... Live DB already updated via direct SQL. This ensures future DB rebuilds and new deployments use ollama as the default provider.	2026-04-30 04:46:04 +00:00
Celes Renata	0665cef7e3	fix: set TZ=America/Los_Angeles in Helm config for all pods ... Containers default to UTC. The host is PDT but pods don't inherit the host timezone. Adding TZ to the ConfigMap ensures all services log and compute timestamps in Pacific time.	2026-04-29 17:53:38 +00:00
Celes Renata	4e010bc048	feat: signal math upgrade — probabilistic, regime-aware scoring pipeline ... Implement full probabilistic signal processing pipeline gated behind probabilistic_scoring_enabled feature flag in risk_configs: - Bayesian log-likelihood accumulator with Beta posterior and entropy - Regime detector (trend-following, panic, mean-reversion, uncertainty) - Source accuracy tracker with per-source historical prediction accuracy - Sigmoid confidence gate replacing binary gate - Information gain surprise weighting for rare events - Adaptive recency decay with event-specific half-lives - Regime multiplier replacing market context multiplier - Weighted disagreement entropy for contradiction detection - Multiplicative macro exposure with conditional integration - Graph-distance attenuated competitive signal propagation - Exponentially weighted momentum with volatility scaling - Expected value recommendation gate All changes backward-compatible: flag=false preserves exact current behavior. New outputs stored in existing JSONB columns (no schema changes except source_accuracy table via migration 034). Tests: 26 property-based tests (14 correctness properties), 99 unit tests, 1789 total tests passing with zero regressions.	2026-04-29 11:41:48 +00:00
Celes Renata	cfcfd655e7	fix: reduce max_tokens to 2048 to fit 8192 context window	2026-04-29 06:18:26 +00:00
Celes Renata	aaf8cee927	feat: scale extractor to 8 replicas	2026-04-29 05:54:00 +00:00
Celes Renata	f264e924f0	fix: thesis rewriter now routes to vLLM when provider is vllm ... - thesis_llm.py: add _call_vllm_thesis() using /v1/chat/completions - thesis_llm.py: check resolved model_provider and route accordingly - values.yaml: set OLLAMA_BASE_URL to http://10.1.1.12:2701	2026-04-29 05:42:10 +00:00
Celes Renata	a36702e5f3	fix: restore REDIS_PASSWORD and MINIO credentials	2026-04-28 22:02:50 -07:00
Celes Renata	01d77c153d	fix: set POSTGRES_PASSWORD in core secrets	2026-04-28 21:59:30 -07:00
Celes Renata	8d227b62f6	scale extractor to 4 replicas	2026-04-28 21:54:38 -07:00
Celes Renata	b38fb24f14	fix: ensure production uses DB-configured model/provider from UI ... - Migration 026: update seed defaults from ollama to vllm/AxionML - Migration 031: fix existing rows still on old ollama defaults - Helm values: set OLLAMA_BASE_URL to cluster ollama endpoint (was empty) - Extractor: guard against switching to ollama when base_url is empty - OllamaClient: validate base_url on construction to fail fast	2026-04-29 04:33:21 +00:00
Celes Renata	23f2134754	fix: stop migrations from resetting ai_agents model_name, default to vllm/AxionML/Qwen3.5-9B-NVFP4	2026-04-28 18:58:13 +00:00
Celes Renata	3b22f5e1fc	feat: seed default risk_configs with macro and competitive layers enabled	2026-04-28 16:34:40 +00:00
Celes Renata	e360b66c3e	fix: beta trading pipeline — max_tokens default, approval re-enqueue, credentials ... - Migration 031: change ai_agents/agent_variants max_tokens default from 32768 to 4096 (32768 exceeds vLLM context window, causing HTTP 400 on every extraction) - API: re-enqueue approved orders to broker queue — previously approved orders sat in DB with nothing to execute them - values-beta: enable TRADING_ENABLED, update Alpaca paper keys	2026-04-28 14:13:58 +00:00
Celes Renata	4bee7a7874	fix: update vLLM model to AxionML/Qwen3.5-9B-NVFP4	2026-04-23 19:30:22 +00:00
Celes Renata	6909ac5e50	feat: add vLLM config to beta values overlay	2026-04-23 08:19:03 +00:00
Celes Renata	117b693b19	feat: add remote vLLM support with provider abstraction layer ... - LLMClient Protocol for provider-agnostic inference - VLLMClient for OpenAI-compatible /v1/chat/completions API - LLM client factory with provider routing (ollama/vllm) - VLLMConfig with VLLM_* environment variable loading - Updated extractor worker with health check and provider switching - Updated event classifier to use LLMClient protocol - Helm values for vLLM configuration - 18 unit tests + 6 property-based tests - Full backward compatibility preserved	2026-04-23 08:17:23 +00:00
Celes Renata	5dcbd286e8	fix: update beta Alpaca broker credentials	2026-04-21 16:43:54 +00:00
Celes Renata	d407888caa	feat: set beta Ollama model to qwen3.6	2026-04-21 09:55:52 +00:00
Celes Renata	abdf639925	feat: isolate beta on stonks_beta database, stop sharing stonks_paper	2026-04-21 09:50:22 +00:00
Celes Renata	a99aaa29ca	feat: point beta Ollama to 192.168.42.254:11434	2026-04-21 09:36:03 +00:00
Celes Renata	08b93ad119	fix: add beta Alpaca broker credentials for separate paper account	2026-04-21 05:32:13 +00:00
Celes Renata	19f39301d3	fix: skip superset/trino ingress when hostname is empty (fixes Kargo beta promotion)	2026-04-21 05:09:08 +00:00
Celes Renata	9275351279	fix: beta stage uses paper DB to resolve permission denied errors ... Beta was pointing at stonks_beta DB where tables were owned by postgres superuser, causing permission denied for the stonks app user. Switch to sharing stonks_paper DB/user (already has proper grants). DEPLOY_STAGE=beta still isolates Redis keys and MinIO buckets. Added market data API key so beta can test ingestion when pipeline is toggled ON.	2026-04-21 04:57:29 +00:00