The engineering questions that matter are rarely about which tool. They are about which failure you are willing to own, and how loudly it will tell you when it happens.
This piece walks through how we think about it at Paycux, what we have changed our minds about, and where the sharp edges are.
First, understand how passwords actually get cracked
That brings us to first, understand how passwords actually get cracked. Make the boundary explicit. When one part of the system can only talk to another through a named interface, you can change either side without a meeting; when it cannot, every change becomes a negotiation.
Measure before you optimise, then measure the thing users feel rather than the thing that is easy to instrument. A p50 that looks fine while the p99 is unusable is a reporting failure, not a performance one.
The entropy math: why length wins
The entropy math: why length wins deserves its own treatment. Idempotency is not a nice-to-have in any system with retries. Key on an identifier the caller supplies, store the outcome, and return the same answer to the same key rather than doing the work twice.
Measure before you optimise, then measure the thing users feel rather than the thing that is easy to instrument. A p50 that looks fine while the p99 is unusable is a reporting failure, not a performance one.
- Name the boundary before you cross it
- Measure what users feel, not what is easy to instrument
- Make every retried operation idempotent
- Write down the assumption that would invalidate the design
What NIST SP 800-63 actually says
What NIST SP 800-63 actually says deserves its own treatment. Measure before you optimise, then measure the thing users feel rather than the thing that is easy to instrument. A p50 that looks fine while the p99 is unusable is a reporting failure, not a performance one.
Idempotency is not a nice-to-have in any system with retries. Key on an identifier the caller supplies, store the outcome, and return the same answer to the same key rather than doing the work twice.
Requirements (SHALL)
Consider requirements (shall). Measure before you optimise, then measure the thing users feel rather than the thing that is easy to instrument. A p50 that looks fine while the p99 is unusable is a reporting failure, not a performance one.
Make the boundary explicit. When one part of the system can only talk to another through a named interface, you can change either side without a meeting; when it cannot, every change becomes a negotiation.
Make the boundary explicit.
Prohibitions (SHALL NOT)
Consider prohibitions (shall not). Measure before you optimise, then measure the thing users feel rather than the thing that is easy to instrument. A p50 that looks fine while the p99 is unusable is a reporting failure, not a performance one.
Measure before you optimise, then measure the thing users feel rather than the thing that is easy to instrument. A p50 that looks fine while the p99 is unusable is a reporting failure, not a performance one.
Why complexity requirements backfire
Why complexity requirements backfire deserves its own treatment. Idempotency is not a nice-to-have in any system with retries. Key on an identifier the caller supplies, store the outcome, and return the same answer to the same key rather than doing the work twice.
Measure before you optimise, then measure the thing users feel rather than the thing that is easy to instrument. A p50 that looks fine while the p99 is unusable is a reporting failure, not a performance one.
Stop forcing password rotation
Consider stop forcing password rotation. Make the boundary explicit. When one part of the system can only talk to another through a named interface, you can change either side without a meeting; when it cannot, every change becomes a negotiation.
Measure before you optimise, then measure the thing users feel rather than the thing that is easy to instrument. A p50 that looks fine while the p99 is unusable is a reporting failure, not a performance one.
Where this leaves us
The pattern repeats across every system we have looked at: the hard part is not the mechanism, it is keeping the mechanism honest as the surrounding assumptions change.
If you are working through the same problem and want to compare notes, the docs cover the mechanics and the console shows the behaviour on your own data.
Everything here, already built
Sign-in, enterprise SSO, directory provisioning, roles and an audit trail behind one API. Start with the quickstart and have a working sign-in this afternoon.