Raul Pacheco Rodriguez
I work to make intelligent, efficient, traceable and reproducible technology useful in the places where large language models cannot reach.
Research direction
Machine learning, quantization, RTL and reproducible medical AI.
2026
Doctoral sprint
FP32
Editorial lab
HW
Aware AI
Collaborations and opportunities
I am looking for scientific collaborations, research opportunities, conferences, specialized teaching and a limited number of technical consulting projects where efficient AI and rigorous systems are central.
Learning AI
I have been accepted into the second intake of the Public Center for Artificial Intelligence Training (CPFIA) to participate in one of the certified Artificial Intelligence courses offered by INFOTEC and private-sector partners such as Google, Microsoft, IBM, Intel, Oracle, or Meta.
New blog
FP32 is the space where I test ideas, technical series, and drafts that may later evolve into books, courses, or research material.
FP32
Essays, papers, research notes and lessons on deep learning, medical AI, efficient systems and tools for augmented researchers.
Research notebook
Short notes on papers, technical decisions, experiments and questions that are still taking shape.
Read notes ->Initial notes on temporal shortcuts, information leakage, and more honest evaluation protocols for medical AI systems.
A practical tutorial-style article about building a personal link hub with HTML and CSS instead of depending entirely on a third-party tool.
A legacy overview of the generative AI wave around 2022 and the questions it opened for creative tools, productivity, and technical education.
Technical library
Readings, repositories and references that feed my research, writing and programming practice.
View resources ->El espacio separado para ensayos, notas largas, series técnicas y futuros libros. El sitio personal funciona como hub; FP32 vive como publicación propia.
Una lista curada de blogs personales, laboratorios y recursos sobre machine learning. Es una referencia útil para organizar lecturas sin convertirlas en posts largos.
Selected work
Experiments, repositories and technical pieces connecting software, hardware and applied AI.
View projects ->Acelerador de la operación softmax para reducir costos de inferencia en modelos de visión profunda como los bloques Shifted Window Transformers.
Rol: Diseño digital, Verilog, FPGA y deep learning
Trabajo en colaboración orientado a mejorar sistemas de clasificación de tejido benigno y cancerígeno mediante modelos de aprendizaje profundo.
Rol: Modelado, evaluación y arquitectura ML
An EdTech platform that transforms corporate knowledge into training through simulated pathways.
Rol: Development
¿Me conviene correr IA local? Descubre si tu equipo actual es capaz de ejecutar modelos de lenguaje (LLMs) localmente, qué rendimiento esperar y si financieramente tiene sentido comparado con pagar APIs como OpenAI o Anthropic.
Rol: Diseño y Desarrollo, Pipeline de datos