Simplifying Quantum Development with Circuit Functions
Key Points
- The talk introduces KKit’s circuit functions and application functions, which aim to give quantum developers higher‑level abstractions similar to those long enjoyed in classical software development.
- Unlike classical programming, today’s quantum programming still requires low‑level work with gates, circuits, and hardware characteristics, forcing developers to manage hardware details directly.
- The quantum software stack sits on top of the physical quantum computer and includes layers such as control systems, error‑handling (correction, suppression, mitigation), and a “primitive” layer that lets developers work with abstract inputs/outputs without manual translation.
- Circuit and application functions handle the necessary transpilation and error‑management steps, turning virtual circuit representations into hardware‑compatible instructions and thus freeing developers from deep quantum‑physics and hardware knowledge.
Full Transcript
# Simplifying Quantum Development with Circuit Functions **Source:** [https://www.youtube.com/watch?v=yjKpfqcnoxk](https://www.youtube.com/watch?v=yjKpfqcnoxk) **Duration:** 00:05:52 ## Summary - The talk introduces KKit’s circuit functions and application functions, which aim to give quantum developers higher‑level abstractions similar to those long enjoyed in classical software development. - Unlike classical programming, today’s quantum programming still requires low‑level work with gates, circuits, and hardware characteristics, forcing developers to manage hardware details directly. - The quantum software stack sits on top of the physical quantum computer and includes layers such as control systems, error‑handling (correction, suppression, mitigation), and a “primitive” layer that lets developers work with abstract inputs/outputs without manual translation. - Circuit and application functions handle the necessary transpilation and error‑management steps, turning virtual circuit representations into hardware‑compatible instructions and thus freeing developers from deep quantum‑physics and hardware knowledge. ## Sections - [00:00:00](https://www.youtube.com/watch?v=yjKpfqcnoxk&t=0s) **Bridging Quantum Development with Abstractions** - The speaker explains how circuit and application functions introduce higher‑level abstractions that simplify quantum programming, contrasting them with the low‑level, hardware‑focused approach traditionally required. ## Full Transcript
hello I'm sank Panda product manager for
the kkit function service today I want
to talk about how Quantum circuit
functions and application functions make
it easier than ever to get useful
results out of quantum computers so
before we talk about circuit functions
and application functions I want to talk
a little bit about classical development
so if you've ever developed with a
classical
computer you are these days You're
benefiting a lot from abstractions right
the your method of development is high
abstracted what I mean by that
specifically is that you can focus on
developing your software rather than
focusing on you know lower level
characteristics such as Assembly
Language machine code the actual
Hardware you're running on you focus
solely on software and less on
Hardware now on the other hand for
Quantum Computing and where it is today
Quantum developers haven't had the exact
same luxury so for a Quantum developer
this
development is still quite
lowlevel what I mean by that
specifically is quantum development is
still happening at lower levels and
representations like circuits Gates and
things like Hardware have to be
intimately considered in the development
of algorithms and application today I'd
like to walk through a little bit of
what does that Quantum soft sofware
stack look like and how circuit
functions and application functions make
it easier to develop
applications so let's start at the very
bottom naturally this is hardware this
is the actual quantum computer that you
can run programs against this does MO
all your computation now as you can
imagine we have a few layers
of software above that that we won't
talk about today but this could include
things like you know control systems or
scheduling so on and so
forth what I do want to start at though
from a software perspective is a
primitive layer at the heart of it I as
a developer can focus on you know
passing in inputs and outputs to that
being able to focus just solely on those
aspects mean that I don't have to you
know translate to whatever lower level
of Hardware language that the quantum
computer
on now given that today's quam computers
are noisy and error prone you also have
some notion somewhere in here of error
handling this can come in a couple of
different flavors you know error
correction error suppression error
mitigation and happens you know all
across the stack but you know when we
talk about getting our circuits our our
observables ready for actual computation
we have some also some notion of trans
transation all this means is you know
translating this virtual representation
or abstract representation of these uh
circuits and observables to something
that is you know interpretable by the
lower levels of our software stack and
our
Hardware now here's where uh you can
probably start telling that to develop
on a quum system you need to have
intimate knowledge of quantum physics
Hardware characteristics selecting what
great transpilation passes you might
use that's not necessarily a skill set
that every developer has so we're
introducing circuit functions as a
concept here as something that I can
focus H change color on just my inputs
you know the quantum
circuit and
observables
so zbl there we go and get at the end of
it mitigated expectation values and
counts with this lovely thing is that
researchers can focus specifically on
creating great representations of their
workloads in that Quantum circuit and
observable thing in that form rather
than worrying about you know
transpilation how does how to get the
maximum out of Hardware characteristics
so on and so
forth now that's one layer of
abstraction this this this circuit
function I want to also talk about one
more layer of abstraction that we have
for application functions now as you can
imagine if I am
a uh computational chemist for example
my goal is to be able to have this uh
like simulate a molecule right I want to
be able to pass in things like a domain
specific input like uh molecular
definition and get out something like
you know energies as a domain specific
output I can take that
information and you know use it for
higher levels of abstraction such as
catalysis or reaction path analysis so
on so forth right but this notion of
mapping my classical
information to Quantum
you know circuits and observables is
also deep research today to see how IBM
and this partners are working on
creating these layers of abstraction I
encourage you to check out some of the
links in the description below to find
out about circuit functions and
application functions thank you so much