Merchly: Risk platform.
A web app for risk & compliance, built with React, Django, AWS and Python. Signals are scored against configurable rules and land in a reviewable queue, so decisions are consistent and recorded.
Selected clients
Power to LiveRBM GlobalDragonflySquiblerMOOSainsbury'sMPCX-LinkPower to LiveRBM GlobalDragonflySquiblerMOOSainsbury'sMPCX-LinkPower to LiveRBM GlobalDragonflySquiblerMOOSainsbury'sMPCX-Link
The challenge.
- Risk signals were scattered, so issues surfaced late and inconsistently.
- The client wanted to score and act on risk from one view.
- Rules had to be adjustable without a code change every time.
- False positives had to stay manageable, or reviewers would start ignoring the queue.
Our approach.
- Built the analyst surface with React for scoring and case review.
- Backed it with Django to ingest signals and store assessments.
- Modelled configurable rules so scoring adjusts without redeploys.
- Surfaced flagged items in a reviewable queue.
- Tuned thresholds with reviewers so the queue stays actionable, not noisy.
The outcome.
- Risk is scored and reviewed from one queue instead of scattered signals.
- Rules adjust without an engineering release.
- Decisions are recorded and auditable.
- Reviewers work a queue they trust instead of tuning out alerts.
Where AI fits in financial services.
Retrieval over policies, filings and statements, assistants for analysts and operations teams, and document checks that leave an audit trail a regulator can follow.
Explore AI developmentAs well as being exceptional technically, they consistently invest the time to understand the product and suggest how it can be enhanced and improved.
Let's build from here.
Thirty minutes with an engineer who builds. No sales, no drip campaign. If we're the wrong fit we'll tell you and point you somewhere better.


