PLATFORM

The Neith Control Layer

One control layer that understands a trade, applies deterministic policy before execution, and records the decision. Five steps, one representation, one record.

WHY NEITH

Trading has execution, risk and surveillance
What it lacks is a control layer

Trading has execution infrastructure. It has risk systems. It has surveillance. What is missing is a control layer that understands trading intent before execution — regardless of where that intent originates. Electronic orders already pass through structured controls. Voice, chat and RFQ trading often does not.

ONE CONTROL LAYER · EVERY TRADING WORKFLOW

VoiceChatRFQAPITrading systemsAutomated trading
↓

NEITH CONTROL LAYER

Capture›Normalize›Policy›Decision›Evidence
↓
Existing execution infrastructure→Market

Neith connects through browser or API, evaluates the trade against controls you configure, records the decision, and passes it on. Today it operates in advisory / shadow mode and does not interfere with execution.

01 · CAPTURE

Read the intent

Read trade intent from a browser or an API, wherever the trade is agreed.

02 · NORMALIZE

One representation

Resolve voice, chat, RFQ and structured input into a common trade record.

03 · CONTROL

Apply policy

Run firm-configured deterministic rules — the same input always gives the same result.

04 · DECIDE

Clear or flag

Produce a clear pre-trade decision in plain English, with low latency.

05 · EVIDENCE

Record it

Write the trade, rule, result, response and configuration to an append-only log.

ONE CONTROL LAYER, END TO END

Control. Decide. Evidence. Integrate.

One architecture, four responsibilities — the same for every desk, channel and system.

CONTROL

Policies, thresholds, permissions and decisioning — configured to your desk and changed without touching infrastructure.

DECIDE

A deterministic rule engine returns a clear pre-trade decision with low latency. Same input, same result, every time.

EVIDENCE

Every decision, policy state and response written to an append-only, tamper-evident record you can export in one click.

INTEGRATE

Two ways in — a browser extension for a low-friction start, or a REST API for direct integration.

Today Neith operates in advisory / shadow mode — it never sits in the execution path.

THE HARD PART IS THE INPUT

Most FX never arrives as structured data.

A voice trade is a sentence; a chat trade is shorthand with fields missing; an RFQ is a screenshot. None of it is a FIX message — which is exactly why static, hard-coded controls miss what happens on this flow. Neith resolves all of it into one structured trade before any check runs, and flags genuinely ambiguous fields on the record rather than guessing.

“mine 10M cable 3M fwd BARC”
↓
BUY · GBPUSD · 10.0M · 3M FWD · BARCLAYS

Two ways in. A REST endpoint if your OMS can call it, or a browser extension that reads the trade where it is agreed — the chat window, the RFQ platform, the front end already open on the desk. A pilot needs no integration project.

WHERE THE MODEL SITS — AND WHERE IT DOESN'T

Models read and explain. Rules decide.

MODEL
Reads the tradeResolving free text, voice transcripts and broker shorthand into structured fields.
MODEL
Writes the explanationTurning a fired rule into a sentence a trader can act on, instead of an error code.
RULES
Make every decisionEach threshold comparison, each flag, each pass. Deterministic and configured by you.

A model cannot raise a flag, and it cannot clear one — every decision traces to an exact rule you configured. That is what makes the record defensible when someone asks, eighteen months later, why a trade was or wasn't flagged.

WHY NOT THE EXISTING APPROACH

Controls bolted to each system
versus one control layer

THE EXISTING APPROACH

✕On voice, chat and RFQ, the control is a person’s attention — and next-day reconciliation
✕Controls built separately around each trading system
✕Voice, chat and RFQ often sit outside structured controls
✕Ownership of controls “poorly defined and not documented” — the regulator’s own finding
✕Every change means integration work across teams and firewalls
✕Evidence assembled after the fact, at reconciliation

WITH NEITH

✓An automated check at the point of the trade, before it leaves the desk
✓One control layer across desks, channels and systems
✓Reads voice, chat, RFQ and structured input alike
✓Every decision owned, attributed and recorded as it happens
✓Change a threshold or policy without rebuilding infrastructure
✓Evidence generated with the decision, not reconstructed later

COVERAGE & CONFIGURATION

Configured to your desk,
not to a fixed list

Neith is not a menu of supported pairs. Every check, threshold and scope is set against your desk's own mandate — the instruments you quote, the tenors you run, and the counterparties on your own approved list.

INSTRUMENTS

Spot, forwards and FX swaps

Outright spot, outright forwards, FX swaps, short-dated rolls and broken dates — including multi-leg trades, validated leg by leg and as a whole.

THRESHOLDS

Tuned per desk and per instrument

A G10 spot book and an EM forwards book run entirely different configurations behind the same endpoint. Notional bands, price collars, tenor limits, settlement rules and approved counterparties are all yours to set — and every change is written to the record.

PAIRS

Any pair your desk quotes

Controls run against whatever your desk quotes. Price checks reference your existing rate feed, and adding coverage is a configuration step against that feed — not a rebuild.

The control layer is asset-class agnostic — price, notional, counterparty, settlement, credit and integrity are not FX-specific concepts. It is designed to extend across asset classes, trading channels and increasingly automated trading environments.

See what Neith would have caught

One desk, six weeks, shadow mode — no execution risk. Five places this quarter.

Start a shadow-mode pilot →