Introducing Complead. One AI-native platform for financial crime compliance. Read the story
New Ready Integrations available Check the new integrations
Complead
Explore free tools Talk to an expert
Agentic AI · Nova

Nova Suggestions you can audit.

Nova reads your screening profiles, rules, customers and monitoring schedules, and comes back with suggestions that carry their own evidence. It proposes; your team approves, rejects or defers. It never changes a setting on its own.

See Nova on your data How the suggestions work
Ask Nova

Ask in your own words, get the evidence back

Nova reads the scan, states what it found and what it could not confirm, and proposes the next step. Nothing is applied until you say so.

Every part Of the score, named
Linked To the record, not restated
Proposes Never applies on its own
Kept Every suggestion logged
Why this exists

Tuning is the work nobody has time for

Screening profiles and monitoring rules are set once, at go-live, by people who are no longer in the room. They are then left alone because nobody can prove in advance what a change would do.

Opaque

a number with no parts

When the score cannot be decomposed, the review becomes a conversation about trust.

Untuned

nobody dares touch it

Parameters set at go-live stay there, because changing them is unmeasured risk.

Undocumented

the reasoning is verbal

The rationale for a threshold lives in a meeting nobody minuted.

How it works

Detect, benchmark, explain, hand over

Scroll to advance
  1. Step 01 Detectors read what already happened

    Deterministic detectors walk your screening profiles, rule firings, dispositions, customer scores and monitoring schedules. No model decides anything at this stage — these are countable facts about your own data.

  2. Step 02 Your history and the sector size it

    A finding only matters if it is out of line. Nova compares each signal against your own baseline and against anonymised sector benchmarks, and drops the ones that are ordinary.

  3. Step 03 Each one is written up with its evidence

    What was found, why it was flagged, which records prove it, how confident Nova is and what it would do to your false-positive volume. Everything an approver needs, on one card.

  4. Step 04 Your team approves, rejects or defers

    Nova stops here. An approval opens a draft rule or a pending setting change for your normal review path; a rejection carries its reason back so the same suggestion does not return unchanged.

nova / detect
  • Alerts read14,206
  • Rules checked38
  • Customers21,480
  • Signals raised96
deterministic detectors · your data only.
nova / benchmark
  • Signals in96
  • Above baseline41
  • Above sector24
  • Kept16
own baseline + anonymised sector medians.
nova / explain
  • Findings16
  • With evidence16
  • Confidence scored16
  • FP estimated11
evidence + rationale + confidence + expected impact.
nova / hand over
  • Approveddrafts only
  • Rejectedwith a reason
  • Deferredreturns later
  • Loggedall three
maker-checker · nothing applied automatically.
Detect
Benchmark
Explain
Hand over
In your workflow

Nova sits in the screen your team already works in

No second tool and no copy-paste. Nova opens beside the case with the context already loaded, and hands the decision back to the analyst.

What it looks at

Four products, one queue

01

Score decomposition

Every contribution shown with its module, its weight and its source.

02

Evidence references

Findings link back to the record rather than restating it.

03

Threshold proposals

A suggested value with the measured effect on volume, never applied on its own.

04

Suggestion history

What was proposed, by whom it was accepted or rejected, and when.

05

Original language kept

Source text is retained alongside any translation.

06

Works inside the case

No separate tool: the answer sits next to the decision it informs.

Guardrails

Nothing is applied automatically

Approve, reject or defer — and every one of those is written down. Rejections carry the reason back to Nova so the same suggestion does not return unchanged.

Nova Suggestions 4
Maker-checker 3
Audit 1

Approve — it becomes a reviewable draft

An approved suggestion opens as a draft rule or a pending setting change, with the evidence attached. It still goes through your normal change process.

Rule Confidence 100% Rule-based

Why Nova flagged this

Nova has write access to drafts, never to live configuration.

Evidence

  • outcome draft created
  • applied by your approver
  • audit event written
Approve · Reject with a reason · Defer — all three are logged
Agentic AI

The rest of the agent suite

Agentic AML-Risk Ops Agentic AI Agentic AML-Risk Ops Screening and monitoring alerts, triaged before anyone opens them. Agentic Fraud Ops Agentic AI Agentic Fraud Ops Transaction alerts scored and routed on your own rules. Agentic SAR Filings Agentic AI Agentic SAR Filings The filing drafted from the case trail, not from memory.
FAQ

Before you switch Nova on

Does the agent act on its own?

Only where you let it. Every action is a node you place yourself, and you can leave a rule in dry run indefinitely.

What if the AI is unsure?

Confidence is a number you can branch on. Most teams auto-close above a high threshold, route the middle band to an operator and escalate the rest.

Can we see what it did?

Every execution writes a log entry: which rule, which node, what it saw, what it changed, and whether it applied, simulated, skipped or failed.

How do we roll it back?

Set the rule to draft and it stops immediately. The log stays, so an auditor can still read what happened while it was on.

Let Nova look at your own queue

We run it against a sample of your screening and monitoring history and walk through what it found.

Talk to an expert Talk to Sales
Minutes To first rule
No code To change one
Dry run Before it touches a case