What is your architecture actually holding up?
TIS Partners runs fixed-scope architecture reviews for systems that carry real load. Two weeks, a written findings report, and decision records your team keeps. Everything after is scoped from what the review finds.
Briefings, by subject
The questions we ask inside engagements, published with the reasoning attached.
Scaling10
Design8
- Is event sourcing worth it?
· 3 min
- How big should a service be?
· 2 min
- What does a decision record owe you?
· 2 min
- What should a system forget?
· 3 min
What do you need looked at?
Three kinds of engagement, each scoped in writing before it starts.
◮
Architecture review
The fixed-scope first engagement. We trace the load path, read the incident history, interview the people on call, and write down what would have to be true for the system to survive its next size.
◭
Scaling engagements
When the review finds a bottleneck worth removing, we design the change with your engineers in the room: capacity models, migration sequencing, and the rollback story before the rollout.
◪
Technical due diligence
For acquirers and investors: what the codebase actually is, what it costs to keep, and which risks are priced wrong. Written for people who sign things.
How does the review work?
The first engagement is the same for everyone: a fixed-scope architecture review. Bounded, priced up front, and useful even if we never meet again.
01
Week one: evidence
Architecture walkthroughs, incident history, dependency and data-flow tracing, interviews with the engineers who carry the pager. We read the diagrams, then we check them against production.
02
Week two: findings
A written report: what is load-bearing, what is fragile, what is merely unfashionable. Ranked by risk, costed in effort, argued in plain language with the evidence attached.
03
Decision records
Every recommendation ships as a decision record your team owns: context, options, tradeoffs, and the question that reopens it. The review is designed to be executable without us.
Asked in the first call
+Do you rewrite systems?
Rarely, and never as the first move. Most systems we review need three targeted changes, not a rewrite. When a rewrite genuinely is the answer, the report says so and explains what the last team that tried learned.
+Our diagrams are out of date. Is that a problem?
It is the normal case. Part of week one is producing an accurate picture; stale diagrams are themselves a finding, and a common one.
+What does the review cost?
A fixed fee quoted after a half-hour scoping call, based on system count and access. The scoping call is free and occasionally ends with us saying a review would not help yet.
+Can our team run the findings without you?
That is the design goal. Decision records are written for your engineers, not for our follow-on sales. Some clients hire us for the next step; some send a note a year later saying what they shipped. Both count as wins.
Start with the question, not the contract.
Describe the system and what worries you about it. You get a straight answer on whether a review would help, and a fixed quote if it would, within two business days.