Machine Learning Engineering
Designing practical ML workflows, model evaluation strategies, and implementation paths that connect data, models, software, and deployment constraints.
Armando Anaya
Artificial Intelligence Engineer
I am a software engineer working with machine learning, computer vision, applied AI systems, robotics perception, and practical AI software.
About
I work at the intersection of software engineering, machine learning, and computer vision. My focus is not only on building models, but on turning AI capabilities into reliable systems that can be integrated into real products, workflows, and research environments.
My academic background includes graduate studies in Computational Engineering and Mathematics at Universitat Rovira i Virgili and Applied Artificial Intelligence at Tecnológico de Monterrey. I also participate in applied computer vision and robotics projects, where technical decisions must balance model performance, data quality, software architecture, and real-world constraints.
Work
The common thread is applied AI: models, data, software, and the constraints that appear when a system leaves the notebook.
Computer Vision
Building and integrating computer vision systems for classification, detection, industrial inspection, applied research, and robotics perception.
Applied AI Software
Developing AI-enabled software systems where models are part of a larger product, service, workflow, or decision-support process.
Robotics and Physical AI
Exploring how perception, simulation, and intelligent systems connect with robotics, autonomous systems, and real-world physical environments.
My approach
I prefer practical, technically grounded AI work. Before building a model, I focus on understanding the problem, the data, the constraints, and the way the solution will be used.
For most projects, the goal is not to use the most complex model available. The goal is to build a system that is useful, measurable, maintainable, and aligned with the real context where it will operate.
-
Start with the problem, not the model.
-
Validate assumptions early.
-
Measure performance with meaningful metrics.
-
Design for integration, not only experimentation.
-
Keep the solution realistic for the available data and infrastructure.