AI Engineer specialized in building and shipping production AI for financial platforms — fraud detection, risk scoring, LLM-powered copilots, and agentic systems that reach real users. I turn ambiguous business problems into reliable, deployed systems, not demos. My work lives at the intersection of applied Machine Learning and real-world finance. At a crypto trading firm I designed and built the company's AI infrastructure on AWS, and shipped a suite of production systems: ML/DL models for fraud and anomaly detection protecting ~1,000 active users; an NLP system that flags advisor misconduct over advisor–client conversations; a financial copilot running on custom LLMs (llama.cpp) with RAG memory; and a behavioral risk-scoring engine for user investment planning. Earlier, I built fine-tuned models, sentiment analyzers, and a custom agent-orchestration library (before adopting LangChain/LangGraph). What I bring: end-to-end ownership — data pipelines, feature engineering, model development, MLOps/CI-CD, deployment and monitoring. Core stack: Python, PyTorch, scikit-learn, RAG, fine-tuning, vector databases, FastAPI, Docker, AWS. I also do the unglamorous engineering that keeps systems alive: CPU-oriented optimization to cut GPU dependency, plus telemetry and profiling to keep production stable. What makes me different: most AI engineers do either generalist LLM work or financial risk — I do both, and I've shipped them. That combination of applied AI + fraud/risk detection + production ownership in fintech is rare, and it's where I'm strongest. Open to remote AI/ML Engineer roles. Let's talk — reach me here on LinkedIn or at _Email not available. Sign in: https://www.twine.net/signup_…AI Engineer specialized in building and shipping production AI for financial platforms — fraud detection, risk scoring, LLM-powered copilots, and agentic systems that reach real users. I turn ambiguous business problems into reliable, deployed systems, not demos. My work lives at the intersection of applied Machine Learning and real-world finance. At a crypto trading firm I designed and built the company's AI infrastructure on AWS, and shipped a suite of production systems: ML/DL models for fraud and anomaly detection protecting ~1,000 active users; an NLP system that flags advisor misconduct over advisor–client conversations; a financial copilot running on custom LLMs (llama.cpp) with RAG memory; and a behavioral risk-scoring engine for user investment planning. Earlier, I built fine-tuned models, sentiment analyzers, and a custom agent-orchestration library (before adopting LangChain/LangGraph). What I bring: end-to-end ownership — data pipelines, feature engineering, model development, MLOps/CI-CD, deployment and monitoring. Core stack: Python, PyTorch, scikit-learn, RAG, fine-tuning, vector databases, FastAPI, Docker, AWS. I also do the unglamorous engineering that keeps systems alive: CPU-oriented optimization to cut GPU dependency, plus telemetry and profiling to keep production stable. What makes me different: most AI engineers do either generalist LLM work or financial risk — I do both, and I've shipped them. That combination of applied AI + fraud/risk detection + production ownership in fintech is rare, and it's where I'm strongest. Open to remote AI/ML Engineer roles. Let's talk — reach me here on LinkedIn or at _Email not available. Sign in: https://www.twine.net/signup_WWWW…