Hire a software architect for the decisions you can't undo
Build versus buy, one service or ten, the data model everything else sits on. Our architects have made these calls before and lived with what each answer cost. You interview them before anyone starts.
What you are getting
4 things that decide this
- 01Senior architects only, each with systems in production that carry the decisions they made, good and bad.
- 02You interview every architect before they join. Nobody is assigned to your project from a profile you never saw.
- 03Every architectural decision gets written down with its trade-off, so the reasoning survives past the person who made it.
- 04They argue from operational cost and reversibility, not from whatever pattern is popular this year.
What a software architect actually does here
The job is not drawing boxes and arrows once. It is deciding what is cheap to change later and what is not, then spending the early weeks on the second kind.
On TrialTriage, the architecture had to hold patient records, trial data and drug data in one place. It also had to support single-patient lookups and insurer batch files, with a nurse reviewing every AI-generated match. That meant a data model built around an audit trail from day one, tracking 23 distinct action types. A compliance gap found after launch is a rebuild, not a patch.
On ZhoopZhoop, the call was which parts of a two-way voice system to build and which to buy. Twilio routes the audio and Deepgram handles speech. The reasoning layer is ours, because that is the part that changes every time a shop changes its process. On Elevent, the decision was how far one WebSocket game engine could scale before it needed to split. The answer was further than most teams assume: the same architecture runs events from 10 players to over 1,000 with no rewrite.
Three different problems, one repeated skill: naming which decision is expensive to reverse, and building the rest so it can change.
What they take off your roadmap
Build versus buy, argued in writing
A short document naming what each option costs to run and to reverse, not a slide deck. The team can point back to it a year later and see why.
The data model that outlives the first release
Schemas and boundaries chosen for what the business will ask of them in a year. Not only what the first feature needs.
A service boundary drawn where it earns its cost
Splitting a system adds a network call, a deploy pipeline and a failure mode. They split where that cost buys something back, and leave the rest as one deployable unit.
A decision record your team can act on without them
The trade-offs get written down alongside the choice. A new engineer six months in can see why the system looks the way it does.
- Scoping callFree. The decision behind the decision
- ShortlistArchitects matched to the problem shape
- You interviewYour process, your bar
- EmbedYour repo, your standups, your reviews
- Decisions loggedWritten down, with the trade-off
The interview is yours. A vetted profile from a marketplace skips the step that actually predicts fit.
Systems built on decisions that had to hold
TrialTriage
AI clinical trial matching for oncology nurses and insurers.
Read the case study →
ZhoopZhoop
AI receptionist and parts procurement for a multi-branch auto repair business.
Read the case study →
Elevent
Real-time multiplayer trivia platform for corporate events and training.
Read the case study →
How hiring works
- 01
Tell us what is stuck
A free call about the system, the constraint and the decision you are stuck on. If the answer is simpler than an architect, you will hear that on the call.
- 02
Meet the architects
We shortlist people who have made a comparable call before, and you interview them. Say no and we go back to the shortlist.
- 03
They embed
Your repo, your standups, your review process. One architect owns the decisions and is named as accountable for them.
- 04
They hand over
The decision record, the diagrams and the reasoning behind each trade-off, plus your team trained to extend it. The first 2 months of support and maintenance are free, with every build.
Stack
Backend and data
Infrastructure
Practices
Tell us the decision you can't afford to get wrong
Bring the system, the constraint and the deadline. The scoping call is free, and you will leave it knowing which decisions are actually expensive to reverse.
01How is this different from a freelance architect or a job board hire?
You get someone who has already lived with a comparable decision in production, plus a company accountable if the call turns out wrong. A job board hands you a resume and a hiring risk you carry alone. If our architect is not right, we replace them, which is not a conversation you have with a contractor you found yourself.
02Does the architect write code, or only diagrams?
They write code. An architect who never touches the codebase produces decisions nobody can implement as intended. They embed with your engineers, build the hardest first pieces, and hand over a system that already runs, not a slide deck describing one.
03How much overlap do we get with our working day?
A daily overlap with your hours, fixed before anyone starts. Architecture decisions get made in conversation, in reviews and in standups, not in a document handed over once a week. We agree the window during scoping and it does not move without you.
04Who owns the decision record and the diagrams?
All of it is yours from the first commit. That includes the decision log, the data model and the reasoning behind each trade-off. It is what lets your team extend the system without calling us back for context.
05What actually drives the cost of an architecture engagement?
How much of the system already exists, and how well it is understood. A greenfield build is mostly decisions; a system already in production adds the work of reading what is there before changing it. Scoping calls are free. Where we have to get into an existing codebase before answering, a paid two-week diagnostic produces a fixed price rather than a guess.
Read next
- monolith vs microservices, the actual trade-off →When splitting a service earns its cost, and when it doesn't.
- build vs buy software →The decision behind the decision architects get hired for.
- AI consulting →Where architecture meets an AI system that has to run in production.
- custom software development →Us owning the build the architecture decisions feed into.
