Summary
Dmitri Leroy is a Senior Software Engineer based in New York who specializes in building production-grade AI platforms that combine LLMs, autonomous agents, RAG, and document intelligence. With four years of professional experience, he brings hands-on expertise across model fine-tuning and GPU-backed inference (PyTorch, LoRA, vLLM), distributed backend design, and high-throughput API systems. He designs for scalability, fault tolerance, observability, and cost efficiency, and frequently wires Redis streams, vector databases, and Prometheus into reliable production workflows. Past roles span end-to-end AI product delivery—from design-to-code pipelines at Quest to operating agentic LLM services at E-Ring—while his open-source work reflects a strong full-stack background in React, TypeScript, and Node.js. Notably, Dmitri pairs deep infrastructure know-how (CUDA-aware optimization and concurrency) with pragmatic frontend and CI/CD practices, enabling seamless model-driven features under real-world traffic.
4 years of coding experience
4 years of employment as a software developer
Political Science and Government, Political Science and Government at Petro Mohyla Black Sea National University
Ukrainian, English