Skip to content
Vates
VAI

The intelligence layer inside Vates

VAI — Vates Artificial Intelligence — reads the market continuously at a scale no team can match, and reports what changed, why it changed, and which of your positions it touches. It does not trade, and it does not decide.

Meet VAI

Hi, I’m VAI — Vates Artificial Intelligence. I’m the intelligence inside the platform.

My job is to take your desk from market data to market intelligence. I read price, flow, filings, and language at a scale a team can’t, and I tell you what changed and why it matters.

I help your team by

  • Watching every instrument on your list continuously, not just the ones in the news
  • Reading filings, transcripts, and coverage the moment they publish
  • Flagging when a correlation you depend on starts to break down
  • Ranking what changed by how much it should matter to your positions
  • Explaining every signal in terms of the data that produced it

You stay in control. I surface what changed and why — the position, the sizing, and the call are yours.

How it works

Read everything. Judge in context. Show the evidence.

Three properties that separate a model worth acting on from one you learn to mute.

01What it reads

Everything, continuously — not just what is trending

Coverage is where most desks silently lose. VAI watches the whole list at the same depth, so a quiet name getting interesting is noticed at the same moment a loud one is.

  • Ticks, volume, order-book pressure, and derivatives positioning
  • Filings, earnings, revisions, and disclosures parsed on publication
  • News, transcripts, and public discussion read for tone and novelty

02How it decides

Baselines per instrument, not one rule for the whole market

A 3% move means something completely different in a rates future and a small-cap. VAI learns what normal looks like for each instrument and measures departure from that.

  • Per-instrument baselines that adapt as an instrument's own regime changes
  • Signals ranked by likely relevance to your positions, not by absolute size
  • Confidence stated explicitly, including when the evidence is thin

03How it explains

Every signal traces back to the data behind it

An alert you cannot interrogate is an alert you will eventually ignore. VAI shows its work, so you can disagree with it on the evidence rather than on instinct.

  • The inputs that triggered a signal, kept alongside the signal itself
  • Plain-language explanation of what changed and against what baseline
  • A full history you can audit after the fact, including the misses
Boundaries

What VAI does, and what stays with you

An honest account of the division of labour. The model handles scale and consistency; judgement, sizing, and execution remain yours.

VAI does not

  • Place, size, or execute trades
  • Make investment recommendations
  • Act on your positions without you
  • Train shared models on your proprietary research
  • Present a guess as a certainty

VAI does

  • Watch every instrument on your list, continuously
  • Learn what normal looks like for each of them
  • Flag departures and explain them against the evidence
  • Rank what changed by relevance to your book
  • State its confidence, including when it is low
One platform. The whole surface.

See Market Intelligence in action

In a demo we will show how Vates ties price, flow, filings, and narrative to the positions you actually hold — and how desks use that to spot regime shifts, crowding, and correlation breaks earlier.