Position: Machine Learning Developer Counterfeit Detection & Risk Scoring Location: Remote/Hybrid Department: AI & Data Science Employment Type: Full-Time About the Role We are seeking a highly skilled Machine Learning Developer to design and implement ML models and pipelines for counterfeit detection, risk assessment, and social media analysis. This role will focus on developing NLP, computer vision, and graph-based learning models to detect fraudulent activities, misinformation, and counterfeit products across digital platforms. The ideal candidate should have experience working with OSINT data, social media monitoring, e-commerce fraud detection, and cybersecurity intelligence. If you have expertise in deep learning, data mining, and risk modeling, we would love to hear from you. Key Responsibilities 1. Data Collection & Preprocessing (optional) Develop web scraping pipelines to collect data from social media, e-commerce platforms, and online forums. Implement EC2-based scrapers for different platforms. Store and preprocess raw data using AWS S3/Google Cloud Storage. 2. Community Analysis & Graph-Based Detection Design GraphDB-based models (Neo4j) to analyze relationships and identify fraudulent social media accounts. Detect fake accounts using similarity measures such as Jaccard Index and account behavior patterns. 3. NLP & Text Data Processing Build BERT/RoBERTa/SBERT-based models to analyze text data for: Brand mentions and anomalies in product descriptions. Misinformation detection and sentiment analysis. Develop rule-based engines for: Brand authenticity verification (e.g., price anomalies, unapproved seller detection). Domain name squatting detection using Levenshtein distance. 4. Computer Vision & Image Processing Implement image preprocessing (resizing, noise correction). Use YOLO-based object detection to verify brand logos and product placements. Develop CLIP-based models for detecting counterfeit images. Implement OCR (Tesseract/EasyOCR) to extract brand names, serial numbers, and other key details. 5. Risk Scoring & Fraud Detection Design a risk scoring system using: Machine learning models (XGBoost, Random Forest, SHAP for feature importance). Heuristic & rule-based fraud detection. Historical fraud pattern analysis. Implement Elasticsearch/SOLR for real-time fraud detection queries. 6. Deployment & Optimization Deploy ML models using Docker/Kubernetes for scalable inference. Optimize models for low-latency predictions in a production environment. Work with MySQL/PostgreSQL for enriched data storage and retrieval. Requirements Must-Have Skills Machine Learning & Data Science: Strong experience in NLP (BERT/SBERT), computer vision (YOLO, CLIP), and fraud detection models. Programming Skills: Expertise in Python (TensorFlow/PyTorch, OpenCV, Transformers, Scrapy). Data Engineering: Experience with AWS S3, Neo4j, Elasticsearch, MySQL/PostgreSQL. Graph-Based Learning: Experience with GraphDB (Neo4j) for community analysis. Risk Scoring & Feature Engineering: Understanding of rule-based and ML scoring models. Deployment & Infrastructure: Experience with Docker, Kubernetes, Flask/FastAPI for serving models. Good-to-Have Skills Experience with OSINT data and cybersecurity intelligence. Familiarity with dark web monitoring tools. Prior experience in social media analytics and misinformation detection. Knowledge of adversarial ML and fraud prevention techniques. Work Mode & Compensation Work Mode: Remote/Hybrid Compensation: Competitive salary + performance-based incentives Growth Opportunity: Work on cutting-edge AI-driven brand protection solutions Job Types: Full-time, Permanent Benefits: Paid sick time Work from home Schedule: Day shift Monday to Friday Morning shift Experience: Machine learning: 3 years (Preferred) Work Location: In person,
Employement Category:
Employement Type: Full timeIndustry: IT Services & ConsultingRole Category: Not SpecifiedFunctional Area: Not SpecifiedRole/Responsibilies: Machine Learning Developer Counterfeit