22 Commits

Author	SHA1	Message	Date
Celes Renata	d9110d03a6	fix: use di.document_id instead of dir.document_id in evidence query ... document_impact_records has intelligence_id, not document_id. The document_id lives on document_intelligence.	2026-05-01 04:06:55 +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	6880f11c26	fix: add /no_think inline tag to disable Qwen3 thinking mode ... chat_template_kwargs isn't being respected by the vLLM deployment. Qwen3 models support /no_think as an inline suffix in the user message to disable thinking mode. This is the most reliable method across all serving backends (vLLM, Ollama, SGLang).	2026-04-29 16:11:44 +00:00
Celes Renata	eead4f1381	fix: disable thinking mode on vLLM path with chat_template_kwargs ... The thesis rewriter uses vLLM (not Ollama) in production. The previous fix only added think=False to the Ollama payload. For vLLM's OpenAI-compatible API with Qwen3 models, thinking mode is disabled via chat_template_kwargs: {enable_thinking: false} in the request body.	2026-04-29 16:04:04 +00:00
Celes Renata	007189c0a5	fix: handle plain-text thinking blocks and disable think mode ... The model outputs 'Thinking Process:' as plain text (not in <think> tags). Updated _strip_thinking_block to handle both XML tags and plain-text reasoning patterns. Also: - Added rule 7 to system prompt: 'Do NOT show your thinking process' - Set think=False in Ollama payload to disable Qwen3 thinking mode - Added fallback regex to extract thesis from after thinking blocks	2026-04-29 15:50:49 +00:00
Celes Renata	f9ee1532dc	fix: strip <think> reasoning blocks from thesis LLM output ... Qwen3.5 in thinking mode emits <think>...</think> chain-of-thought before the actual response. The thesis rewriter was returning the raw output including the entire reasoning block. Now strips thinking tags from both Ollama and vLLM response paths.	2026-04-29 15:25:04 +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	8c3c1aab43	fix: pipeline stop now halts all workers and flushes queues ... Workers (ingestion, parser, extractor, aggregation, recommendation, broker, lake-publisher) now check the pipeline:enabled Redis flag on each loop iteration and sleep when disabled. The toggle endpoint flushes all pipeline queues on disable so queued jobs don't resume when workers eventually check. Broker/trading queues are excluded from flush to avoid dropping in-flight orders.	2026-04-29 07:59:35 +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	a6189fce6b	fix: skip LLM thesis rewrite for informational/suppressed recs to prevent queue buildup	2026-04-17 19:29:41 +00:00
Celes Renata	7736554988	fix: recommendation worker now respects thesis-rewriter active=false toggle	2026-04-17 19:22:31 +00:00
Celes Renata	419cf7558a	fix: evidence articles missing on recommendations + Lucide title prop CI failure ... - recommendation worker: filter out non-UUID document IDs (synthetic pattern:* IDs from competitive signals) before inserting into recommendation_evidence table — the uuid cast was failing and silently dropping all evidence rows - wrap executemany in try/except so partial failures don't lose all evidence - SqlExplorer: wrap Lucide icons in <span title=...> instead of passing title prop directly (not supported by lucide-react, broke CI build)	2026-04-17 07:10:21 +00:00
Celes Renata	7c23c044d7	feat: agent variants — migration, API, service integration, frontend, tests ... - Migration 027: agent_variants table with single-active enforcement, variant_id column on agent_performance_log - API: full CRUD, clone from agent/variant, activate/deactivate, per-variant performance metrics and history endpoints - Services: extractor, event classifier, thesis rewriter all wired to AgentConfigResolver with variant override support - Frontend: variant list, comparison view, create/edit/clone forms, activate/delete actions on Agents page - Tests: API tests + 5 property-based tests (single-active invariant, clone preservation, config resolution, slug determinism, update idempotence) - Spec files for agent-variants feature	2026-04-17 05:15:42 +00:00
Celes Renata	6179382d1e	feat: wire all 3 agents to DB config resolver ... - Recommendation worker now resolves thesis-rewriter config from DB and passes ollama_config to generate_recommendation. Thesis rewriting is now active when the thesis-rewriter agent exists in ai_agents. Refreshes config every 50 jobs. - Event classifier now resolves its own config separately from the document extractor via 'event-classifier' slug. Uses a separate OllamaClient when the model differs from the extractor. Refreshes alongside the extractor every 100 jobs. - Document extractor was already wired (existing code). - Added 8 unit tests for AgentConfigResolver covering: DB resolution, variant override, not-found, DB errors, TTL caching, cache refresh, and invalidation.	2026-04-17 02:59:40 +00:00
Celes Renata	f11aa0a1ee	fix: deduplicate recommendations and widen position sizing range ... - Add dedup check in recommendation worker: skip generation when latest rec for same ticker+window has identical action/mode/confidence - Widen position sizing range (1-10% portfolio, 0.3-2% max loss) and factor in trend strength + evidence count for differentiated sizing - API returns only latest recommendation per ticker by default (DISTINCT ON) to eliminate duplicate rows in the frontend list view	2026-04-17 00:15:32 +00:00
Celes Renata	d8ea58104c	fix: lint errors (import sorting, unused vars)	2026-04-14 19:48:19 +00:00
Celes Renata	f7a11d14ea	feat: competitive intelligence & historical pattern matching layer	2026-04-14 19:42:48 +00:00
Celes Renata	b478022ba3	fix: data quality query and suppression fallback in recommendation worker ... - Fix _DATA_QUALITY_QUERY: remove nonexistent d.source_id/s.source_class, use d.source_type directly - Fix LIMIT 1 applied after jsonb expansion by restructuring as CTE - Fix fallback build_quality_context_from_summary returning empty source_types which always triggered LOW_SOURCE_DIVERSITY suppression - Update test to reflect corrected fallback behavior	2026-04-14 06:57:46 +00:00
Celes Renata	d16e15c885	phase 17: quote reserved word 'window' in all SQL queries across recommendation worker and query API	2026-04-12 03:45:51 -07:00
Celes Renata	109440c91e	phase 15: fix ruff lint errors across services	2026-04-11 12:10:01 -07:00
Celes Renata	ce10afa034	phase 14-15: docker build validation and helm deployment	2026-04-11 11:59:45 -07:00
Celes Renata	ebea70573b	phase 0+1: project scaffold, k8s manifests, CI pipeline, steering, hooks, tests ... - Repository structure for all services, infra, lakehouse, dashboards - K8s manifests targeting stonks-oracle namespace with GHCR images - Ingress via Traefik with ca-issuer TLS for internal services - ConfigMap wired to existing cluster services (pg, redis, minio, ollama) - GitHub Actions workflow for lint, test, multi-service container builds - Dockerfile with build-arg CMD per service - Makefile for local build/push/deploy - Steering rules for TDD workflow, K8s conventions, project context - Agent hooks for lint-on-save, test-on-save, k8s-validate, phase-commit - Ruff linter config, all lint issues fixed - 14 passing tests for schemas, config, redis keys - PostgreSQL migrations, Trino catalogs, Superset config, MinIO lifecycle	2026-04-11 03:25:08 -07:00