Tomás
Short-form Insights
One question leads to another. One optimization exposes a new limitation. Along the way: caching, TTLs, stampedes, coalescing, Redis, and shared failure—not as a tech checklist, but as the natural consequence of the same constraints.
A browser asks for a website. The answer is an IP address—but getting there means asking the right servers in the right order. Along the way: resolvers, roots, TLDs, nameservers, A/CNAME/MX/NS records, TTLs, and Anycast—not as a glossary, but as the natural consequence of one question.
Greedy algorithms make locally optimal choices at each step. They work when the problem has optimal substructure and the greedy choice property—when choosing the best option now doesn't prevent you from finding the global optimum. Most of life doesn't work this way, which is why intuition fails us in complex systems.
You can't have Consistency, Availability, and Partition Tolerance all at once. Pick two. In distributed systems, as in relationships, you're always trading off between staying in sync, being present, and handling disconnection gracefully.
Sapir-Whorf for programmers: writing Haskell makes you think in types and transformations. Writing Python makes you reach for pragmatic solutions. The patterns you internalize become the lens through which you see problems.
Curated Notes
Abelson & Sussman
Still the best introduction to thinking computationally. Not about Lisp—about abstraction, recursion, and how to build complex systems from simple parts.
Martin Kleppmann
The modern bible for understanding distributed systems. Every chapter is a rabbit hole worth exploring. Read it twice.
Marcus Aurelius
A Roman emperor's private journal on dealing with chaos, uncertainty, and the gap between intention and action. Two thousand years later, still relevant.
Projects that caught my attention
I wanted to manipulate objects with my hands—abstractly—and let any app listen to those events. The canvas was the lab; the iteration is what made the signal trustworthy enough to share.
View →My first project when I started with HTML, CSS, and JavaScript. A simple game—mostly an excuse to make something real and see if the code would actually run.
View →I'm most excited by difficult technical problems with meaningful human impact.
I'm happy to work outside my current specialization when the problem is worth solving.
If that means learning robotics, embedded systems, computer vision, AI, or another field — that's part of finding a way.
I care about using AI to democratize access — and about its failure modes, privacy, and what we should be careful building.
Books that shaped my thinking
Douglas Hofstadter
The book that made me understand recursion, self-reference, and why consciousness might emerge from strange loops.
Carlo Rovelli
Time doesn't flow. It's layered. Rovelli writes physics like poetry.
Nassim Nicholas Taleb
Some things benefit from disorder. Applies to systems, careers, and life strategies.
Jorge Luis Borges
Short stories as thought experiments. The Library of Babel changed how I think about infinity.
Currently
Reading: Designing Data-Intensive Applications
Last run: 5K this morning
Languages
Cat person
Working on a weird idea, a hard technical problem, research, or something that doesn't exist yet?
I'd love to hear about it — especially open-ended work where technology can meaningfully help people.
Let's figure out what it would take.