Hi, I'm Braddy

Futureproofing Tines: Fair share orchestration

Thu, 29 Jan 2026 12:00:00 +0000

Originally published on the Tines Engineering Blog.

One of the more fun and challenging problems I worked on at Tines - redesigning how we orchestrate action runs so that no single customer’s workflow can starve everyone else. With over a billion automated actions flowing through the platform every week, our initial simple approach needed an upgrade. We built a fair-share system that pairs deterministic concurrency gating with continuously accruing per-story budgets, and the result is much more predictable performance across the board.

Making N+1 query detection actually useful at scale

Thu, 01 Jan 2026 12:00:00 +0000

Why N+1 queries matter

An N+1 query is one of the most common performance anti-patterns in applications powered by relational databases. It happens when code loads a collection of records, then lazily fetches an association for each one individually effectively turning what should be 1 query into N+1.

# 1 query to load users
users = User.where(team_id: team.id)

# + N queries, one per user, to load their profile
users.each { |u| puts u.profile.bio }

At large scale where N might be 100, 200, or 1,000, it’s not a great thing:

Pairity

Wed, 05 Nov 2025 00:00:00 +0000

From a player’s frustration to builder

For a few years now, I’ve been a member in various badminton clubs in Ireland. Each club has their own way of doing matchmaking, whether that is using traditional peg boards where players manually track the queue and pairings, or use software like BadBoard which promised automation. The process is tedious and repetitive and either method does not do it reliably nor efficiently. It led to the same people sitting out, or unfair skill levels, or repeated matches, with no insight into how it’s picked or match history.

Some advice to my younger self

Sat, 04 Oct 2025 12:00:00 +0000

Build your own worldview

It is important to develop the skill and habit of thinking for yourself. This aligns choices with values which reduces regret, second-guessing, and a more focused growth. This also allows you to also resist trends and borrowed opinions. All in all, it’s clear standards and integrity, high agency, and cleaner tradeoffs.

Avoid generalization from specific experiences

It’s very easy to default to generalization from specific experiences, especially when contexts are not static. When listening, actively look for examples that weaken the theory and then find ways to improve on the theory. This way, we can understand the world deeply by improving our explanations.

Truths about software engineering I learned

Mon, 08 Sep 2025 12:00:00 +0000

After withdrawing from ETH Zurich in a degree for MSc in CS, majoring in Theoretical Computer Science and Machine Learning, I joined a startup in pursuit of making impact, learning with practical experience, and building. Here is what I learned.

The power of dogfooding and observability

It’s much easier to come up with ideas when using the product as you feel the rough edges firsthand which promotes thinking about ways to improve it. This is similar to instrumenting what users feel, you can’t fix what you can’t see.

Four disciplines in shipping what matters

Fri, 01 Aug 2025 12:00:00 +0000

Recently at work, I started a recurring monthly session focusing on growing engineers, especially younger engineers. These sessions are mostly focusing on career advice, principles, and tools to make engineers more effective.

This session focused on how to increase impact and demonstrating that impact.

A small disclaimer before we continue, (good) advice is fuzzy because it entirely depends on the person receiving it.

Four disciplines

In software engineering, there are four disciplines:

The pen is mightier than the AI model

Wed, 23 Jul 2025 12:00:00 +0000

Today, I received the results and feedback of my product improvement submission for an internal transfer to a PM role. Although writing the submission was already a deeply fulfilling and fun experience, the discussion I had after with Head of Product transcended the experience and led to an important lesson and a reality check for myself.

In pursuit of my idea of “perfection”, I employed the help of AI to aid my storytelling techniques, challenge my assumptions, and guide my answer. However, I failed to realise that the more I used AI to critique and refine, the more I lost the human touch - the clarity of thought and voice.

You don't have to choose between infra and product

Sun, 08 Jun 2025 12:00:00 +0000

Early in my career, I started out in a product-facing team, working on user-visible features, collaborating with design, thinking about experience and usability. It was rewarding, fast-paced, and full of feedback loops.

After a while, I knew I wanted to push myself further.

I didn’t just want to design interfaces. I wanted to understand the systems beneath them. I wanted to write code that was resilient under load, optimize for throughput, and work at scale. I wanted to build the kind of technical judgment that only comes from shipping real infrastructure.

Speed is a feature and why moving fast improves quality

Sun, 01 Jun 2025 12:00:00 +0000

“Speed and quality aren’t at odds - they’re often positively correlated.” - Nan Yu, Linear

When I first started out as an engineer, I assumed that “moving fast” meant rushing, cutting corners, or skipping due diligence. Sure, sometimes it does if you’re careless. However, over time, especially working in a startup environment, I’ve learned something that surprised me:

Speed can actually improve quality.

Not in a reckless way, but in a focused, practiced, high-feedback-loop kind of way. The best engineers and teams I’ve worked with move quickly because they care about quality. They know what matters, what can wait, and how to learn through iteration. They don’t aim for perfect upfront. They aim to learn fast and improve continuously.

About me

Thu, 21 Mar 2024 00:00:00 +0000

I’m Braddy, a senior software engineer at Solidroad, a YC-backed AI QA platform for enterprise CX. I work across the stack, shipping customer-facing product as readily as the distributed infrastructure beneath it, and spent the previous four years at Tines delivering new revenue channels, leading the infrastructure for a product-led-growth launch, and scaling multi-tenant orchestration to over a billion workflow actions a week. I like the problems where product and systems constraints collide.

Reading list

Thu, 21 Mar 2024 00:00:00 +0000

While books are valuable resources, the best learning comes from understanding fundamentals and building real things. This is just a complement to hands-on experience.

Books

Distributed systems

Designing Data-Intensive Applications
Database Internals

Math

Statistics Without Tears

Programming

Tidy First?
Refactoring

Sales

Let’s Get Real or Let’s Not Play: Transforming the Buyer/Seller Relationship

Soft skills

Unleashed: The Unapologetic Leader’s Guide to Empowering Everyone Around You

Misc

The Beginning of Infinity: Explanations That Transform the World
Voices of the Fallen Heroes

Papers

Distributed systems