PA.
HomeProjectsTech StackBlog
Resume
PA.

Senior AI/ML Engineer · KYC/AML · Africa

HomeProjectsTech StackBlog

© 2026 Patrick Attankurugu. Built with Next.js.

Patrick Attankurugu

Patrick
Attankurugu

Senior AI/ML Engineer · Financial Compliance AI · Africa

Helping African financial institutions prevent fraud, verify identities, and stay compliant — with AI built for scale.

View My WorkGet in Touch
Chat on WhatsApp · +233 551 087 418
About

Building AI for Africa

Jim Leech Mastercard Foundation Fellow at Queen's University, Canada — one of just 60 selected from 1,000+ entrepreneurs across all 54 African countries. I'm a Senior AI/ML Engineer specialising in financial compliance AI: building systems for Know Your Customer (KYC), Know Your Business (KYB), Anti-Money Laundering (AML), and transaction monitoring.

Currently Senior AI/ML Engineer at Agregar Technologies, architecting identity verification, deepfake detection, credit scoring, and AML transaction monitoring platforms for enterprise financial institutions across Africa.

BSc Computer Science from the University of Ghana. I build for Africa because the continent's financial institutions deserve compliance AI as sophisticated as anywhere else in the world.

Deep Learning

PyTorch, TensorFlow, ResNet, transformers for production inference

Computer Vision

YOLO, OpenCV, document verification, deepfake detection

KYC/AML Compliance

Identity verification, transaction monitoring, PEP screening, FATF compliance

LLMs & Agentic AI

LangChain, LlamaIndex, RAG pipelines, autonomous compliance agents

Skills & Expertise

Technical Stack & Domain Expertise

AI & Deep Learning

PythonPyTorchTensorFlowscikit-learnONNXCNNsTransformersResNetYOLOGANsModel QuantizationTransfer Learning

Computer Vision & Biometrics

OpenCVFace DetectionLiveness DetectionDeepfake DetectionOCR / PaddleOCRDocument VerificationSmart ID CaptureVideo AnalyticsObject DetectionImage Classification

KYC / KYB / KYE

Identity VerificationDocument AuthenticationBiometric MatchingBusiness VerificationUBO DetectionEmployee ScreeningPEP ScreeningSanctions ScreeningWatchlist MonitoringRisk Scoring

AML & Transaction Monitoring

Transaction MonitoringSuspicious Activity DetectionSAR/STR GenerationRule-Based EnginesML Anomaly DetectionFATF ComplianceTypology MappingRisk AssessmentAlert TriageRegulatory Reporting

LLMs, RAG & NLP

LangChainLlamaIndexVector DatabasesRAG PipelinesPrompt EngineeringAgentic AIspaCyEntity ResolutionFuzzy MatchingText Classification

Engineering & Infrastructure

FastAPINext.jsReactTypeScriptPostgreSQLRedisNeo4jDockerAWSCI/CDREST APIsSystem Design
Experience

Where I've Worked

Agregar Technologies

Current

Senior AI/ML & Backend Engineer

Aug 2024 – Present

Building and maintaining the core AI/ML systems that power identity verification and financial compliance for enterprise clients across Africa. Architecting AI credit scoring, fine-tuning ResNet and YOLO models for document verification and smart ID capture, AML transaction monitoring engines, and a deepfake detection system for biometric fraud prevention.

PythonPyTorchTensorFlowResNetYOLOOpenCVFastAPIPostgreSQLRedisDockerAWSLangChain

SEMA Technologies

AI Engineer

Nov 2022 – Jul 2024

Built a real-time predictive surveillance system that interfaces with IP cameras to detect and predict suspicious activity using computer vision and ML models. Led a team of 5 engineers.

PythonPyTorchOpenCVTensorFlow

SuperFluid Labs

AI Engineer · Team Lead

2020

Built an LLM-powered solution with a 4-person engineering team for an AI innovation challenge, finishing in the top 4 of 12 competing teams.

PythonLLMsPrompt Engineering

Kosmos Innovation Center

Technical Lead · AgriTech Challenge

2023

Led technical development for an agriculture technology project integrating data-driven approaches to cassava processing and export optimization.

PythonData AnalysisML Modeling

DEXT Solutions Consult

Web Developer

Jun – Aug 2019

Built websites and web applications for clients. Streamlined codebases and improved user experience across multiple projects.

JavaScriptHTMLCSS
Projects

What I've Built

Featured ProjectLive · Open Source

AfricaPEP

No dedicated PEP screening database existed for African jurisdictions — so I built one. Covers all 54 African countries with 27,000+ verified Politically Exposed Persons (PEP) profiles, an NLP entity resolution pipeline with hybrid fuzzy matching, and FATF-compliant tier classification.

27K+PEP Profiles
54Countries
1,300+Family Links
FastAPINeo4jPostgreSQLNext.jsspaCyrapidfuzzDocker
Live Demo Source Code

SENTINEL — Autonomous Compliance AI

Open Source

Five specialized AI agents orchestrated by LangGraph run end-to-end AML/KYC investigations autonomously — triaging alerts, gathering evidence, and drafting Suspicious Activity Reports (SARs) without manual intervention.

PythonLangGraphGeminiFastAPINext.js

Deepfake Detection System

Enterprise

Production biometric fraud prevention — detects AI-generated faces, video replays, and printed photo spoofing in live KYC deployment.

PythonPyTorchDeep LearningOpenCV

AI Credit Scoring Engine

Enterprise

Machine learning credit scoring system for financial institutions — alternative data analysis, risk modeling, and automated decisioning.

Pythonscikit-learnFastAPIPostgreSQL
View all projects
Achievements

Milestones & Recognition

Jim Leech Mastercard Foundation Fellow

2023

Queen's University, Canada

One of just 60 fellows selected from over 1,000 applicants across all 54 African countries — a leadership programme backed by the Mastercard Foundation for transformative entrepreneurs.

BSc Computer Science

2021

University of Ghana

Transitioned from Chemistry to Computer Science, graduating with a focus on AI and data science.

AI Innovation Challenge — Top 4

2020

SuperFluid Labs & AI Commons

Led a 4-person engineering team to build an LLM-powered solution, finishing in the top 4 of 12 competing teams.

AgriTech Innovation Challenge

2023

Kosmos Innovation Center

Built a data-driven platform to model cassava processing efficiency and export optimisation, leading a cross-functional technical team selected for the programme.

Blog

Insights from the Compliance AI Frontier

Why Rule-Based AML Systems Fail: How ML Changes Everything
AML
March 202510 min read

Why Rule-Based AML Systems Fail: How ML Changes Everything

Most financial institutions still rely on static thresholds to detect money laundering. The result? Thousands of false alerts and sophisticated patterns slipping through. Here's why, and what ML changes.

Deepfake Detection in Production KYC: Lessons from 100K+ Verifications
Computer Vision
January 20259 min read

Deepfake Detection in Production KYC: Lessons from 100K+ Verifications

Research papers report 99% accuracy. Production reality is different. The failure modes nobody publishes, the adversarial attacks customers actually try, and the decisions that matter.

From 10,000 False Alerts to 200: Rebuilding Transaction Monitoring with ML
Transaction Monitoring
November 202411 min read

From 10,000 False Alerts to 200: Rebuilding Transaction Monitoring with ML

When 95% of your AML alerts are false positives, your compliance team isn't investigating, they're drowning. How we rebuilt a monitoring system from scratch.

Building Autonomous Compliance Agents with LangGraph
AI Agents
September 202410 min read

Building Autonomous Compliance Agents with LangGraph

Five specialized AI agents that conduct full KYC/AML investigations autonomously: from alert triage to SAR drafting. The architecture, the guardrails, and what we learned.

View all 10 posts

Get in Touch

Open to new opportunities, collaborations, and conversations about AI/ML engineering. Whether it's a role, a research discussion, or just connecting — feel free to reach out.

Emailpatricka.azuma@gmail.com
WhatsApp+233 551 087 418
LinkedInpatrick-ai-africa
GitHubPatrickAttankurugu
Based in Accra, Ghana

Open to opportunities and conversations worldwide