Andrés Ruiz Calvo

• Architected comprehensive RAG pipelines from ground up, implementing using Python sophisticated data chunking algorithms, optimized Elasticsearch vector indexing, multi-stage retrieval mechanisms, advanced re-ranking with cross-encoder models, context-aware LLM prompting strategies, and intelligent query rewriting for enhanced semantic understanding. Deployed production-grade FastAPI microservices containerized with Docker, orchestrated via GitLab CI/CD pipelines for DevOps integration
• Developing AI agent systems capable of autonomous code generation in Angular, implementing even self corrective steps and unit tests. Utilizing LangGraph while working with diverse agent architectures and tools. Successfully deployed applications on serverless cloud containers with integrated GitHub repository connections. I also exposed the agentic applications with remote MCP servers (hosted on AWS) using fast-mcp
• Initiating advanced web scraping projects (using python, selenium and S3 buckets) to develop robust data ingestion pipelines for enhanced RAG application performance and knowledge base expansion
• Delivered comprehensive workshops to customers covering GenAI concepts including prompt engineering techniques and RAG application development, building technical credibility and client relationships
• Participating in executive communication initiatives with senior company leadership to develop my public speaking expertise

Skills: Python · Elasticsearch · PyTorch · AWS Bedrock · LangGraph · FastAPI · Docker · Git · CI/CD · Agents

Associate Professor in Machine Learning in the AI and Computation Degree at University UAX. I created all the theory and coding labs for the Machine Learning I & Machine Learning II curricula for 20+ students.

Solutions Architect responsible for Spain's hybrid cloud projects using Huawei Cloud Stack (HCS)

• Solutions Architect leader of HCS project sold at 600,000€ to consulting company. I was responsible for the technical solution and communication with their CTO
• Prepared and delivered introductory technical presentations to different small and medium sized companies about HCS
• POC leader with startup focusing on VR and AI workloads, I successfully migrated their CDN and GPUs instances to Huawei Cloud
• Collaborated on a project with a software company to deploy a ML model using Huawei Cloud infrastructure. Working with senior SA, my team is mainly focusing on designing the auto scaling policies for the container instances and implementing a queue strategy
• Currently working on an interactive dashboard (using Python) for internal use that automatically generates insights about billing details
• Prepared a technical proposal for a startup whose main architecture consists of containers and NAT gateways to handle network traffic
• Worked with BDs to create offerings for customers of compute (ECS) and storage services (OBS) in Huawei Cloud
• Obtained the Associate and Professional Solutions Architect Huawei Cloud certifications (fastest one on my team) and created a study guide

Solutions Architect responsible for Spain's hybrid cloud projects using Huawei Cloud Stack (HCS)

• Created a public cloud backup solution for a telecommunication consulting company, including both technical and economical proposal

Universidad Carlos III de Madrid

Relevant Courses: Machine Learning, Deep Learning, Computer Vision, Data Modelling, NLP, Information Theory.

Universidad Carlos III de Madrid

Relevant Courses: Supervised/Unsupervised Learning, Reinforcement Learning, Transfer Learning, Databases, Neural Networks, Deep Learning, Programming (Python,R,SQL, Scikit-Learn), Statistical Learning, Data structures, Big Data.

AMAZON WEB SERVICES

AMAZON WEB SERVICES

DATABRICKS

AMAZON WEB SERVICES

In a practical workshop at PyConES (biggest python conference in Spain), I instructed over 50 participants on how to build a multimodal RAG (Retrieval-Augmented Generation) pipeline from the ground up. The complete solution adhered to industry best practices, using Python and IBM's robust open-source library, Docling, for intricate document processing.

Applied Optics

First-author of a study published in Applied Optics: Developed using python a deep learning segmentation algorithm (feature pyramid network with pre-trained encoder) for anterior eye OCT images, achieving 93.2% accuracy and 0.34 s/image processing.

Lifelong passion for football, tennis, running, and calisthenics, with a commitment to regular practice and personal fitness.

Enjoy reading, particularly autobiographies and reflective literature, to gain new perspectives and insights.

Maintain a personal blog focused on sharing lessons from exceptional individuals through storytelling; publish weekly to an engaged audience of approximately 200 readers.

Enjoy exploring new cultures and destinations, fostering adaptability and a global outlook.