Nicolas Dartayeta

I’m a young software and AI engineer with a solid academic foundation in Software Engineering and hands-on experience building both backend and frontend systems. I specialize in ML & AI capabilities like RAG, agents, and computer vision, with a strong focus on making solutions fast, reliable, and cost-effective.

In my current role, I’m leading the end-to-end development of a greenfield digital vehicle marketplace, architecting core services from the ground up and selecting optimized tech stacks for each component. I designed a real-time chat microservice (Node.js), implemented a Python microservice for automated vehicle listing examinations using locally hosted Small Language Model + Computer Vision, and engineered a secure PHP/MySQL foundation with robust authentication/authorization and scalable schemas. On the UI side, I leveraged Next.js to deliver a responsive, high-performance experience integrating real-time messaging and dynamic listing states.

Previously, I designed and built a digital marketplace from the ground up with a team of 4 developers, using modern tooling like Bun.js, Astro, React, Tailwind, PostgreSQL, Git, and Docker—focused on real-world impact (food waste reduction). I also build advanced ML applications through projects such as a full-stack RAG pipeline (Celery/Redis, Docling parsing, hybrid chunking, streaming LangGraph agents with cited sources), nanoGPT-style transformer training (Optuna, PyTorch Lightning, MLflow), and computer vision and speech-to-text apps using YOLOv11 and Whisper, always with production-minded engineering practices.