The Principles

Principles are a compass, not a GPS

We let the current context, not generic rules from past experiences, guide our decisions. Our mental models inform — they don’t dictate.

Trust your people

We hire great people, define a clear mission and outcome, and trust them to shape the best role to achieve it. We provide context and support, not a rigid box.

Be the navigator, not the captain

Our experienced folks provide the map, the safety net, and shared wisdom — they don’t grab the wheel. Growth comes from walking the rope yourself, knowing someone’s got the net.

Assume positive intent

We start from the belief that people mean well. No question is a "dumb" question. We create space where everyone feels safe to say "I don't know" and to seek clarification with curiosity, not suspicion.

Blame the game, not the player

When a mistake happens, our first question is "What in our process allowed this?" not "Who did this?". Fixing the system provides long-term value; blaming creates fear.

Smooth seas don’t make skillful sailors

True growth comes from navigating challenges, not avoiding them — and from pausing to absorb the lessons. We embrace setbacks as opportunities and make deliberate time to reflect on what they taught us.

Pass the torch

We actively work to make ourselves redundant by documenting knowledge, sharing skills, and empowering teammates. Knowledge silos are a liability; collective resilience is the goal.

Rest is a feature, not a bug

Consistent, high-quality work is the product of a well-rested team. We actively encourage a sustainable pace.

Async-first

We default to writing to protect focus time. Meetings are a high-cost exception — used intentionally for complex collaboration, alignment, and team connection.

Public by default, private for people

Work discussions go in public channels for visibility and shared learning. DMs are for sensitive matters or personal feedback — we protect teammates from being put on the spot.

Communicate with context

We respect our teammates’ time by providing the necessary context upfront. No messages that require three follow-ups just to understand the ask.

Feedback is for growth, not just correction

Every moment of feedback is an opportunity to help each other grow. We don’t just point out what to fix; we explain the why to scale the team’s collective skill.

Challenge ideas, not people

We have strong opinions, weakly held. We listen first, debate the underlying principle when we deadlock, and give each other space to articulate the "why."

Disagree and commit

Once a decision is made, everyone commits to giving it the best possible chance of success. This isn’t blind loyalty — if evidence says we’re wrong, we raise the flag and re-evaluate as a team.

Simplicity is a prerequisite for success

Complexity is our enemy. If a solution — or our explanation of it — isn’t simple, it isn’t mature yet. We do the hard work to reach the simplicity on the other side of deep understanding, and we build the simplest thing that solves the problem.

Develop your taste

Technical skill is knowing how to build something. Taste is the wisdom to know what is worth building and how it should be built for this specific context. We cultivate taste by learning from different approaches and understanding the trade-offs.

Heads up, noise down

We resist tunnel vision and side quests equally. We look ahead to avoid strategic dead-ends, check the rear-view mirror to de-risk accumulated debt, and ignore 90% of the feedback to find the 10% that actually matters.

Know the season. Advocate for the next

We adapt our engineering style to the project's phase — prototyping (spring), reliability (summer), paying down debt (winter). Sometimes we move fast to earn customer trust ("build trust"); at other times we build the robust, long-term foundation ("build forever"). We advocate for the season the product needs next.

Ship and iterate

We are strategic about delivering the highest value in the shortest time. Today’s implementation may be imperfect, but we ship it because it creates value now — with a clear mental map of how we’ll evolve it. We don’t wait for the perfect design; we iterate towards it.

Conscious trade-offs, not shortcuts

A trade-off is a conscious, visible decision with a plan to address the debt. A shortcut is a reactive choice with a vague hope of "we'll fix it later." We don't do gut-feeling engineering.

Ownership doesn’t end at the merge button

We own the bugs, the pager, and the infrastructure of what we ship. Our responsibility runs from ideation through production support.

One change at a time

We never mix migration with mutation. Whether it’s moving from SQLite to Postgres or restructuring a module, we separate the shape change from the behavior change. One PR moves the furniture, the next one redecorates. This keeps work reviewable, reversible, and safe.

Always build an escape path

We don’t prematurely commit to a new path by immediately deleting the old one. We treat deployments as two-way doors — keeping the old code active or togglable until the new way proves its resilience. We only tear down the old path once the new one has fully earned our trust.

Invest in artifacts that stay honest

Tests, schemas, and infrastructure-as-code are the source of truth — because they’re executable and can’t silently drift. Specs and design docs matter too, especially now that they’re the context AI reads to build the right thing. We maintain both: runnable artifacts for correctness, written specs for intent.

The excavator problem

When we all had shovels, the pace of digging gave us time to think and adjust. AI is the excavator — so fast that if you haven’t thought it through before you start, you’ve already dug the wrong hole. We use it to accelerate, but speed demands that the thinking happens before the doing, not during.

We work backwards from the customer

We start with the customer experience and work backwards to the technology. User benefit drives technical decisions, not the other way around.

Evidence precedes execution

We don’t build on assumptions. Every feature needs specific user feedback, data, or a business case behind it. And we’re trained to spot the XY Problem — we solve the root cause, not just the presented symptom.

We own the solution, not just the execution

We are partners, not an order-taking service. Our job isn’t to type what a ticket says — it’s to understand the customer’s problem and make sure what we build actually solves it.

Partner with the accountable

Our most valuable partners are those with real skin in the game. We focus collaboration on them, which lets us filter signal from noise.

Vision leads, expertise informs

The person accountable for the outcome holds the decision. Their job is to seek and deeply understand the forceful counsel of domain experts before making the final call.

Treat trust as capital

We build social capital by consistently creating alignment and delivering value. We spend it consciously, and only on high-conviction decisions critical for our customers or the company.

Speak their language

The same decision is a user-impact story for product, a risk story for leadership, and an architecture story for engineering. We understand what each audience values and frame our message in their terms — because the best idea, poorly translated, dies in the room it needed to win.

Win the war, not just the battle

We differentiate strategy (long-term plan) from tactics (individual actions). We constantly ask: "How does this task serve our larger goal?"

Shorten the loop

Our most valuable asset is validated learning. We prioritize shortening the feedback loop between shipping and understanding impact. The faster we learn from real users, the faster we build what truly matters.

Outcome over output

Our success is measured by real-world impact on users and the business, not by the volume of features we ship. We aren’t here to ship more things to keep it exciting; we are here to ship the right things. Technical purity that doesn’t sell is a failure.