blog: Qiskit IVR functional validation#47
Conversation
ajbozarth
force-pushed
the
docs/qiskit-ivr-functional-validation
branch
from
46613d8 to
a3fe774
Compare
Signed-off-by: Alex Bozarth <ajbozart@us.ibm.com>
ajbozarth
force-pushed
the
docs/qiskit-ivr-functional-validation
branch
from
a3fe774 to
eb50af3
Compare
planetf1
left a comment
planetf1
left a comment
There was a problem hiding this comment.
Nice follow-up to the first IVR post. The strategy-flip under richer feedback caught my attention, and the scope-honesty section about what does and doesn't transfer lands well. One small adapter-experience note inline.
vabarbosa
left a comment
vabarbosa
left a comment
There was a problem hiding this comment.
thank you @ajbozarth very nice!
i just have a couple minor comments below
cbjuan
left a comment
cbjuan
left a comment
There was a problem hiding this comment.
Just a minor nit (not blocking). I like the blogpost (and results). Good work. Thank you, @ajbozarth
- Reframe QHE PR reference as a snapshot ("looked like at the time of this benchmark") so the prose ages well after the PR merge state changes.
- Soften "in my toolbox repo" to "on GitHub" to better match the surrounding tone.
Assisted-by: Claude Code
Signed-off-by: Alex Bozarth <ajbozart@us.ibm.com>
planetf1
left a comment
planetf1
left a comment
There was a problem hiding this comment.
LGTM. I couldn't run anything, but checked for technical correctness.
It's an interesting exploration of how mellea can help, though the sample sizes and scope is very small to make a general conclusion on. It sits best as a pragmatic experience than then data-driven research - which is how it's described.
More examples like this (in future blogs) will be a good thing
5 participants
Summary
check()into the IVR loop as a second validator alongside QKTthe same 302-run benchmark
check()validator is dataset-bound, but the integration pattern transfers to any project's tests
Test plan
npm run lint:mdcleannpm run dev/blogs/qiskit-ivr-code-validation/resolves/schedule 2026-05-27