Private beta

The epistemic trading copilot

Grounded in live market intelligence — not vibes. Trading Galaxy fuses 22 live data sources into a calibrated knowledge base, then reasons over it to give you signals you can actually trust.

Live right now: 10,000+ market intelligence atoms · 22 adapters · updated every minute

No spam. Export to Mailchimp when we launch. Unsubscribe any time.

✓ You're on the list. We'll reach out when beta access opens.
KB atoms live
22 live data adapters
5 min enrichment cycle
on the waitlist

Built differently

Most trading tools give you charts and headlines. We give you structured, calibrated knowledge.

Epistemic grounding

Every signal is backed by a KB atom with a subject, predicate, object, confidence score, and source. No hallucinations — the copilot only asserts what the data supports.

Knowledge Graph
📡

22 live data sources

yFinance, FRED macro, EIA energy, GDELT geopolitical tension, UCDP conflict, ACLED unrest, USGS seismic, SEC filings, RSS feeds, options flow, FCA shorts, and more — all fused into one graph.

Multi-source fusion
🎯

Calibrated signals

Signals come with conviction tiers (High / Medium / Low / Avoid), signal quality scores, macro confirmation flags, and geo-risk exposure — so you know exactly how much weight to place on each call.

Confidence-weighted
🌍

Geopolitical risk layer

GDELT bilateral tension, UCDP active conflicts, and EIA energy data combine to emit geopolitical_risk_exposure atoms per ticker and an energy_shock_risk macro signal — automatically injected into every chat response.

Geo-risk enrichment
💬

Portfolio-aware chat

The AI copilot knows your holdings, cost basis, risk profile, and current regime. Ask "what's my biggest risk right now?" and get a grounded answer, not a generic one.

Personalised context

Causal shock engine

When a macro event fires — an EIA energy shock, a USGS seismic event in an oil region, or a GDELT bilateral tension spike — the engine propagates the impact to related equities automatically.

Causal propagation

How it works

A structured intelligence layer between raw data and your decisions

01

22 adapters ingest live data

Every 1–60 minutes, specialised adapters pull from yFinance, FRED, EIA, GDELT, UCDP, USGS, SEC EDGAR, FCA, RSS feeds, and more. Each datum is normalised into a typed RawAtom.

02

Signal enrichment runs every 5 minutes

A second-order pass reads the raw atoms and emits derived signals — price regime, conviction tier, signal quality, macro confirmation, geopolitical risk exposure, and energy shock risk.

03

The knowledge graph resolves conflicts

Contradictory signals are detected and scored for epistemic stress. Confidence decays over time. The graph always reflects the current, weighted state of evidence — not stale cached values.

04

The copilot answers with grounded context

Your question is matched against the KB. Relevant atoms are retrieved, ranked, and injected into the LLM prompt alongside your portfolio. The answer cites the data it was built from.

Get early access

We're opening the beta to a small group of traders and researchers. Join the list to be first in.

No spam. Unsubscribe any time.

✓ You're on the list. We'll reach out when beta access opens.