Muhammed Keita
ML Systems Engineer · MLOps · FinTech ML · GCP Certified
New to Contra
Muhammed is building their profile!
Cloud-Native Fraud Detection API — FastAPI inference service deployed on AWS EC2 with GitHub Actions CI/CD to AWS ECR, MLflow Model Registry integration, and Kubernetes manifests ready for EKS.
0
24
Credit Card Fraud Detection Pipeline — 13-experiment benchmark across 5 imbalance strategies × 3 models. XGBoost + class weighting wins: PR-AUC 0.880, Recall@P90 0.837.
0
22
Production ML Monitoring API — FastAPI, Docker, GitHub Actions CI/CD, custom statistical drift detection. Monitors deployed classifiers for data drift and model degradation in production.
0
20
Heart Disease Risk Pipeline — end-to-end ML system, Random Forest AUC=0.921, Recall=0.922. MLflow experiment tracking on DagsHub, live inference on Hugging Face Spaces, GitHub Actions CI/CD.
0
14