Charm Security-AI & ML Data Scientist

Description

We are a stealth startup led by experienced founders, backed by Team8, a leading VC firm, building AI-powered scam prevention solutions, and we’ve just secured seed funding! 

About the Role:

We are looking for an AI & ML Data Scientist with expertise in LLM fine-tuning, retrieval-augmented generation (RAG), and fraud detection models. This role is ideal for a data scientist passionate about deploying AI at scale, optimizing real-time fraud detection, and integrating advanced ML techniques into cutting-edge security solutions.

Key Responsibilities:

• Develop and optimize AI-driven fraud models, leveraging LLMs, RAG architectures, and classical ML techniques.
• Fine-tune large language models (LLMs) to enhance scam detection capabilities in real-time environments.
• Build scalable, real-time fraud prediction pipelines for anomaly detection and transaction risk scoring.
• Research and apply graph-based fraud detection techniques and network analysis for identifying fraudulent patterns.
• Optimize AI models for latency, accuracy, and adaptability in high-volume fraud prevention use cases.
• Work with data engineers and software engineers to deploy AI models efficiently in production.
• Stay at the forefront of LLM, NLP, and AI research, implementing the latest advancements in fraud prevention.

Requirements

•  5+ years of experience in Data Science, AI, or NLP, preferably in fraud detection, security, or fintech.
• MSc/PhD in Computer Science, Machine Learning, Mathematics, or a related field.
• Expertise in Python, Classic ML frameworks, and NLP techniques.
• Hands-on experience fine-tuning LLMs and/or implementing RAG architectures.
• Experience deploying AI models in cloud-native environments (GCP/AWS, Kubernetes, or serverless architectures).
• Strong understanding of graph-based fraud detection, anomaly detection, and adversarial ML techniques.
• Experience with vector databases (e.g., Pinecone, Weaviate) and AI-driven search (advantageous).

Nice to Have

• Background in cybersecurity, threat intelligence, or financial crime AI applications.
• Experience with feature stores, data versioning, and MLOps workflows.
• Familiarity with federated learning and privacy-preserving AI techniques.

This is a unique opportunity to be at the forefront of AI-driven fraud prevention, leveraging cutting-edge machine learning, LLMs, and real-time detection to combat emerging threats. 🚀