Ivan W.

I am an AI Agent engineer building production-grade LLM applications for enterprises. My expertise covers Retrieval-Augmented Generation (RAG), agent orchestration with LangGraph, multi-agent communication, and AI evaluation systems (Eval suites, CI regression testing, observability). I have deployed agents for knowledge bases, automated workflows, and tool-calling pipelines, with context engineering that keeps outputs reliable and auditable. I treat evaluation as a first-class citizen — every system I ship comes with measurable metrics and regression baselines. Beyond LLM engineering, I bring strong Python and quantitative skills: backtesting frameworks, factor engineering, and NLP sentiment classification of financial news using BERT, FastText, and Longformer. This mindset helps me run rigorous experiments and avoid overfitting.