Risk / AML / Fraud system for crypto-acquiring (real-time)

Hello,

I would be happy to carry out this project. I have a ready Proof of Concept (PoC) architecture in Python/FastAPI that I can show immediately – it includes a rules engine, scoring, and audit logging.

How I will do it:

1. Backend: Python (FastAPI) or Go – microservices architecture.
2. Logic: Hybrid engine (hard rules + scoring) + Redis for real-time anomaly detection.
3. Integrations: Adapters for AML providers (e.g., Chainalysis) and full JSON audit for compliance.
…
I can issue an invoice as I am on JDG my website www.volexit.pl

Show original

Translate

Hello!
My name is Alexey, I represent a group of developers - NC-1.
In our team, there is a specialist (Backend-developer, Data Engineer, middle) with the necessary experience and knowledge for you. The hourly rate is 68 PLN.
Stack: Transact-SQL, MS SQL Server, SSIS, OLAP, ETL, SQL, Python, Git, Gitlab.
Cases: https://1drv.ms/b/c/b7a0d31a9dae1bc5/IQCpK38gmEvWT6F_Cso40Li-AXAKkSs-J67mCwll-C732pw?e=v4VVF5
Sincerely,
Alexey M.

Show original

Translate

I am ready to complete your project with quality in code and beauty in design - quickly and efficiently. I have extensive experience in similar work. Feel free to contact me.

Show original

Translate

I would create a separate microservice:

For incoming/outgoing transactions, we will implement Chainalysis KYT - this is real-time monitoring of transactions with alerts (they claim literally seconds after the on-chain event).
To check addresses quickly and BEFORE the operation - we will use Address Screening, purely to assess the risk profile of the wallet. And the Sanctions API is a must (based on SDN/OFAC lists), so that sanctions are immediately a hard stop.
Additionally, we will add transactional/behavioral factors: velocity, anomalies in amounts, repeatability, deviations by merchant (conversion/decline rate), novelty of the user, etc.
An important point - we need to run through the history, we will adjust weights and thresholds to fit your reality, so as not to accidentally kill the conversion.

I can suggest assembling a "quick MVP" (minimum rules + Chainalysis + logging). We will fix the effect, and then we will expand the coverage.

Show original

Translate

Hello.

My name is Karen, I am a backend developer and team lead. I work with fintech projects and systems where transaction decisions are made in real time. I am familiar with crypto acquiring and risk logic not theoretically, but practically.

I understand the task as building a transparent decision engine, where each decision can be explained and reproduced. Not a black box, but a managed system, ready for scaling and regulatory issues.

What I propose:

— formalization of risk factors by merchant, user, transaction, and blockchain
… — rule-based layer for strict limitations and stop factors
— scoring model with a single risk score
— flexible configuration of approve, review, reject thresholds
— integration of blockchain AML checks and external providers
— complete logging, decision history, audit trail

Technically:

— backend in Python or Node depending on the current stack
— real-time risk calculation without blocking the payment flow
— separate risk service with the ability to expand logic
— neat documentation for the team and compliance

I am interested in a production result and long-term work, not a one-time implementation. I am ready to discuss the current architecture, data sources, and propose a phased implementation plan.

Karen

Show original

Translate

Hello.

Yes, the task is clear and matches my profile.

I have practical experience in systems that work with real crypto transactions, decision flow, risk logic, auditing, and anti-fraud patterns. I see the task specifically as building an explainable risk engine, not a "black box."

Here is how I envision the solution at a high level:
- Formalization of risk factors (merchant / user / transaction / blockchain)
- Rule-based layer + scoring model
… - Calculation of a unified risk score with an explainable breakdown
- Decision engine (approve / review / reject) with configurable thresholds
- Integration of AML providers (through adapters)
- Complete audit trail (input → score → decision → rule)

The architecture is designed to be modular, with the possibility of further scaling (additional rules, new AML sources, ML layer in the future).

I am ready to discuss:
- Current AML providers (if any)
- Expected false-positive rate
- Ownership decision thresholds
- Timelines and format of interaction

The price of 40,000 PLN seems realistic with a clearly defined scope.

I am ready to move on to discussing the architecture and requirements.

Show original

Translate

Good day, Denis!

Creating a real risk/AML/fraud system for crypto-acquiring is a challenge that I am ready to take on. With experience in developing corporate systems and integrating AML programs, I can offer a unique approach to rules and scoring, while ensuring the transparency of algorithm operations. The main focus will be on formalizing risk factors and integrating AML, which will allow for prompt assessment of transactions. Let's discuss the details and create a reliable system for your business.

Show original

Translate

Hello! I am ready to complete this project and have extensive experience in developing various applications.

Show original

Translate

Good day!
My name is Valentin, and I represent Arctic Web Agency. We are a team of experienced web developers who specialize in creating modern and effective web solutions for businesses. I can provide examples of our similar work in personal messages. We are ready to take your project to work!

Sincerely,
Arctic Web Team
Freelancehunt

Show original

Translate

Hello, Denis

There are some small developments in anti-bot solutions and experience in integrating AML solutions. Also, the development of a risk score for the company's internal audits.

I can start working on this already this month, write to me.

Show original

Translate

Hello, I am a Golang developer with over 4 years of commercial development experience in Go, including in blockchain.
Project examples:
- SaveX - MPC crypto wallet for stablecoins
- faces.foundation - our new startup for KYC tokenization and crypto asset protection
- Granium - decentralized file storage system based on Cosmos SDK
- kyc.market - KYC exchange with crypto payments

For such a project, I propose part-time collaboration with a fixed time/rate. I think we can come to an agreement.

Show original

Translate

Hello,

I have hands-on experience designing and implementing real-time risk assessment systems for payment and crypto-related products, with a strong focus on AML, fraud detection, and merchant risk. I’ve built explainable, rule-based and scoring engines that produce unified risk scores and clear approve/review/reject decisions, with full logging and audit trails.

I’m comfortable formalizing risk factors across merchants, users, transactions, and blockchain data, integrating external AML providers, and deploying production-ready backend systems in Python and Node.js. My approach prioritizes transparency, scalability, and fast iteration as risk logic evolves.

I’d be happy to discuss architecture, timelines, and pricing, and I’m open to long-term collaboration.

Best regards,

Hello!
I have practical experience in designing and implementing risk-based decision-making systems for financial and crypto payment products, focusing on AML, antifraud, and merchant risk. I understand well that in your case, it is not about individual checks, but about a transparent and manageable decision engine that makes real-time decisions to approve/review/reject and can scale.
My approach
I implement a rule-based + scoring model without "black boxes," with explainable logic:
Formalization of risk factors at levels:
merchant (type of business, history, behavior, velocity);
user (recurrence, geography, behavioral patterns);
transaction (amount, frequency, deviation from the baseline profile);
blockchain (source of funds, exposure, hops, sanction risks).
… Calculation of a single risk score with clear weights and reason codes.
Configurable decision thresholds (approve/review/reject) without the need for releases.
Integration of blockchain AML providers through an adapter layer (Chainalysis / TRM / Merkle Science / analogs).
Complete audit trail:
input data;
triggered rules;
score;
final decision;
policy version.
Architecturally
Separate decision engine with API and low latency.
Versioning of policy and rules.
Storage of key features and history.
Logging and metrics for further tuning.
Backend: Python / Go / Node — according to your stack.
Result
A functioning risk system in production;
Documented AML/antifraud logic;
A foundation for further scaling (new rules, providers, merchants).
Questions for a correct start
What blockchains and assets are used (BTC / ETH / TRON / stablecoins)?
Is there already a chosen AML provider?
What KYC/KYB data is available for users and merchants?
Expected load and latency requirements (TPS, SLA)?
What MVP volume do you consider sufficient for the first production launch?
I suggest fixing the timelines after agreeing on the MVP scope.
I would be happy to discuss the details and potential long-term cooperation.

Show original

Translate