Taylor Brooks
Builder of Things
about me
My name is Taylor and I love building things. I've been helping to design and build everything from web apps to data pipelines to engineering teams for over a decade. I'm passionate about building products, processes, and teams to help solve some of the 7,142,853 different problems you may face on any given day in the world of software engineering.
my values
I found a great deal of clarity and self-awareness from the practice of declaring my values. Each of these values is something I strongly believe in. Like all humans, I'm not perfect, but I always try my hardest to incorporate these values into every decision I make.
Integrity
- Our industry today runs on powerful and eye-catching headlines, which all too often are solely backed by half-truths and claims of little substance. To me, Integrity means a dedication to the pursuit and communication of the truth. When possible, decisions should be grounded in evidence. If we built something amazing, but it has a few flaws, we should be proud of what we built AND honest about its flaws (and what we are doing to address those flaws).
- While it might be difficult at times, this dedication to the truth builds a powerful trust between us, our teammates, and our customers. It lets us be open and honest with our teammates, and it gives us the confidence to know we're moving in the right direction.
- We should be consistent in our ideals, but flexible in our decisions. If new data becomes available that conflicts with our earlier assertions, we should be ready and willing to tell our audiences that we were wrong, and share our reasoning.
Empathy
- We all make mistakes. If we're wrong, Empathy allows us to see the other side's argument and realize our mistake.
- If users are willing to take the time to buy and use our products, we should give them the respect they deserve. We should treat them how we would wish to be treated as users. This means, before we build a new feature, we should understand how the user is currently using the product. If time allows, we should prototype that feature and use it ourselves. All my favorite products were designed/built by the same people who use them, and I don't think that's a coincidence!
Simplicity
- It's not always possible to find simple solutions to complex problems, but the constant pursuit of simple solutions leads us naturally to the most pragmatic and cost-effective paths forward.
- A slight increase in complexity leads to a significant increase in problems. Complexity is a quadratic multiplier on the costs and risks of every step of the software development lifecycle. It can force us to forsake paths that would have otherwise led to a more successful, cost-effective, and user-friendly product.
- Beware the "siren's song" of abstraction. Abstraction can alleviate symptoms, but Simplicity is a real cure. For example, creating an in-house devtool that automatically manages developer environments naturally means fewer engineers will understand how their developer environments work. Furthermore, the abstraction creates a justification for those same developer environments to grow in complexity, further exacerbating the problem.
what im doing now
In January of 2022, I joined HashiCorp to work on Terraform Cloud, a service that helps engineering teams and hobbyists across the world manage their infrastructure as code. If you work at a company or even a small side-project that you need to provision infrastructure for, you can try it out for free. I used it to deploy this website!
Most of my free time these days is being spent on (of course) agentic development. Sometimes, I wish AI would disappear off the face of the earth, but I can't deny how powerful it is as a tool. When agentic development first started to pop off, and I started to see all these posts claiming that "software engineering is dead, look what I just vibe coded", I was terrified. That terror led me to actually take a serious look at how transformers work, what are the best ways to use agents... and why almost everything I read on the internet about the power of vibe coding (especially if it comes from someone who is selling AI) was completely false. Getting my hands dirty and building products with all kinds of different tools, harnesses, and models has given me a sense of understanding and empowerment. I've seen the flaws of powerful models like Fable 5 and the power that comes with spec-driven development. I know that while we can reduce the amount of manual verification work that is needed to build products with agents, we still need to check every line that goes to production. I feel I have a clearer vision of what the future of "predictive text generation" will be for the next couple years.
When I'm not working on side-projects, I'm usually spending time with my partner and our two dogs, wood-working, metal-working (thankfully, I'm not as bad at software as I am at welding), riding dirt bikes or traveling.