Diego Rodriguez specializes in AI/ML engineering and production full-stack development. His core focus areas are LLM orchestration (building applications powered by large language models using LangChain and LangGraph), RAG (Retrieval-Augmented Generation) pipeline construction for domain-specific knowledge retrieval, AI agent systems with tool/function calling for autonomous business process automation, and ML-driven risk detection using scikit-learn and behavioral feature engineering. He also builds algorithmic trading systems using reinforcement learning (PPO and SAC algorithms with PyTorch). Diego bridges the gap between cutting-edge AI research and real-world software products that serve paying users in production at scale.

Diego's primary stack combines Python and TypeScript across the full product. On the AI/ML side he uses Python with FastAPI, LangChain, LangGraph, scikit-learn, PyTorch, and Pandas; vector databases including pgvector (PostgreSQL extension), Pinecone, and Chroma; and all major LLM providers — OpenAI GPT-4o, Anthropic Claude, Google Gemini, and open-source models via Ollama or Hugging Face. On the application side he uses TypeScript, React, and Next.js for frontends; Node.js or NestJS for backend APIs; PostgreSQL, MongoDB, and Redis for data storage; and AWS, Docker, Kubernetes, and Vercel for deployment and infrastructure.

Diego has delivered multiple production AI systems. At a high-traffic gaming platform (2024–2025) he built an ML-driven fraud detection and risk scoring engine achieving 94% precision using scikit-learn, behavioral feature engineering, and a real-time FastAPI scoring endpoint. At Proxify (2025) he built an AI-native payment gateway using the x402 protocol, enabling autonomous AI agents to process transactions via tool/function calling interfaces — one of the first production deployments of this emerging standard. At iDelsoft (2025–present) he is building an AI-powered SaaS platform for the US market featuring LLM-driven proposal generation, an intelligent recommendation engine, and payment processing infrastructure.

Yes. Diego is based in Montevideo, Uruguay (UTC-3) and takes on remote freelance engagements with clients worldwide. He operates async-first — meaning clear written communication, thorough documentation, and predictable delivery without requiring real-time overlap. He is available for both project-based work (fixed scope and price) and monthly retainer arrangements (ongoing feature development, AI integration, and maintenance). Typical projects start within 1–2 weeks of agreement. To discuss your project, schedule a free 30-minute discovery call at diego-rodriguez.work/schedule.