requirements to specs
by Adrian Zambrana
Convert natural language requirements into validated OpenAPI contracts, Gherkin features, and AI agent tool specs.
Hi, I'm Adrián 👋 I build AI solutions that work in production, from predictive models to conversational agents powered by LLMs. My focus is on Applied AI & Data Science: turning complex problems into working systems using Machine Learning, RAG pipelines, and agentic architectures. I enjoy the full journey from prototype to deployment. Check out my work at azafuture.com, where I share projects, experiments, and practical AI implementations. What I work on 🤖 AI Agents: conversational systems, RAG pipelines, LangChain, multi-agent architectures 📊 Data Science: predictive modeling, demand forecasting, customer segmentation 🧠 Deep Learning: computer vision (YOLO, ResNet, EfficientNet), neural networks 🚀 MLOps & Deployment: CI/CD, Vertex AI, Cloud Run, Arize, RestAPI, FastAPI, Docker, AWS, Streamlit, Railway 🔄 Automation: Make, n8n workflows, intelligent pipelines About Me 🛸 Full Stack LLM Developer at Accenture 🎯 Applied AI Engineer Previously at Aselvia, building conversational agents with Python, RAG, LangChain, FastAPI and vector databases 📈 Previously Data Scientist at Tiendas MGI demand forecasting, Power BI dashboards, pricing optimization 🎓 Master's in Applied AI from Instituto Europeo de Posgrado 🛠️ I prototype fast, iterate often, and ship things that solve real problems 🎨 Senior 3D Artist background for video games