Introducing Power BI: a 2016 book saved by one chapter A free introductory book written when the product was brand new. Most of what it describes no longer exists under that name. The part that did not age is the data model. Architecture 19 mar 2026
Fundamentals of Software Architecture: there are no right answers Its first law is that everything in architecture is a trade-off. What I kept is not the catalogue of styles but the habit of writing down the alternatives that were rejected, and why. Architecture 19 feb 2026
Fundamentals of Data Engineering: what runs underneath every stage A five-stage data lifecycle and, cutting across all of it, six undercurrents. The second idea is the one I use; the first I already had. Architecture 16 feb 2026
Enterprise Architecture as Strategy: the question that comes before the map Twenty years and a cloud later, its opening question is still the best in its field: how much data integration and how much process standardisation does this particular company need. Architecture Strategy 31 ene 2026
Distributed AI Systems: three chapters for me, the rest for another trade A book about training and serving models on GPU clusters. Not my work and never will be, yet the inference chapters answer the question I actually get asked: what does this cost per month. AI Architecture 31 dic 2025
Database Design: a university text that aged unevenly Nearly three hundred pages of Colombian coursework. One chapter explains something I watch break every week; two others are query drills that were already redundant when it went to print. Architecture 28 dic 2025
Designing Machine Learning Systems: the model is the easy part It lays out the full lifecycle of a machine learning system. The half that aged is the training half; the half that holds up is everything that happens after something goes live. AI Architecture 22 dic 2025
Designing Data-Intensive Applications: where the data is born, and who owns it The book that gives integration arguments a vocabulary. It left me a rule I apply before proposing to connect two systems: draw where each piece of data is born first. Architecture 19 dic 2025
Data Engineering with Python: the tools aged, the checklist did not Building data pipelines with the tools of 2020. I read an extract and do not need more: what survives is the chapter on what separates a toy pipeline from one that runs every night. Architecture 7 dic 2025
Data Acquisition Handbook: why the scale integration fails in the cable A vendor reference manual, unsigned, about turning the physical world into numbers: thermocouples, strain gauges, noise, sampling. I open it when a peripheral returns data nobody can explain. Architecture 4 dic 2025
Cloud Strategy: nobody wants a server, they want to stop thinking about one Thirty-two short essays about deciding on a cloud, not operating one. This is the book to read before quoting a migration, and the one that explains why the promised savings rarely show up. Architecture Strategy 24 nov 2025
Building Microservices: the best argument against microservices Newman warns in the opening chapters that almost nobody needs this, and the industry read the title anyway. For a mid-sized company the usable chapter is a different one: the one about transactions that do not exist. Architecture 18 nov 2025