Your First Step Into Quantum Computing Doesn't Require a Physics PhD
AWS now lets you run code on real quantum computers with no upfront commitment and no physics degree. Here's how Amazon Braket works and what it takes to get started.
Series: Quantum on AWS, Without a PhD (2 articles)
- 1Your First Step Into Quantum Computing Doesn't Require a Physics PhD This article
Most people hear quantum computing and picture a lab full of cryogenic equipment, physicists in white coats, and math that would make your head spin. Fair enough, few years ago, that was more or less the reality. If you wanted to run a quantum algorithm, you needed physical access to a specific machine from a specific vendor, along with their proprietary tools.
That barrier is gone now. Amazon Braket puts quantum hardware in your hands through the same AWS console you already use for EC2, S3, and Lambda. No lab coat required.
I recently took the Amazon Braket Getting Started course on AWS Skill Builder, and it made something clear: the entry point into quantum computing is a lot more accessible than most people think. Here's what I learned.
What Amazon Braket Actually Is
Braket is AWS's fully managed quantum computing service. Through its APIs and Python SDK, you can:
- Run circuits on real quantum processing units (QPUs) from multiple hardware providers
- Test your code on managed simulators before touching actual quantum hardware
- Build hybrid algorithms that combine quantum and classical compute resources
Think of it as a single front door to quantum computing. Behind that door are superconducting processors, trapped-ion machines, and neutral-atom devices, each from different vendors, each with different strengths. Braket gives you one consistent interface to all of them.
The Problem It Solves
Here's the situation Braket was built for. Say you're a researcher or developer who wants to explore quantum algorithms. Until recently, your options looked like this:
- Find a hardware vendor that has available capacity,
- Negotiate access (often a subscription or grant),
- Learn their specific SDK and tooling,
- Hope the machine is available when you need it,
- Start over if you want to try a different vendor's hardware.
That's a lot of friction for technology that's still in its early stages, the stage where experimentation matters most. Braket collapses all of that into a single service. One SDK, one billing model, multiple quantum technologies. You can run the same algorithm on an IonQ trapped-ion machine and a superconducting device, then compare results side by side.
It's also fully integrated into the AWS ecosystem. Your results land in S3. Access is controlled through IAM. You monitor usage with the same tools you use for everything else on AWS.
How It Works: Build, Test, Run
The course breaks Braket's workflow into three stages, and the simplicity is the point.
Build: You start in a managed Jupyter notebook environment. Braket notebooks come preinstalled with the SDK, sample algorithms, and tutorials. You write your quantum circuit in Python using the Braket SDK, and the same code runs on any supported device with minimal changes.
Test: Before you spend money on QPU time, you test on simulators. Braket offers local simulators (free, included in the SDK, good for up to ~25 qubits) and on-demand cloud simulators that auto-scale for bigger workloads. The cloud simulators bill per millisecond of simulation time, with a 3-second minimum.
Run: When your circuit is ready, you send it to a real QPU. Braket queues the task, runs it when the device is available, and stores results in S3. You pay per task (a per-shot fee plus a per-task fee) with no upfront commitment.
The key idea: your code stays the same across all three stages. You swap the target device, not the logic.
Under the Hood
Here's how all the pieces fit together:
You write code using the Braket SDK (or Qiskit, PennyLane, or Julia if you prefer). That code defines a quantum circuit, a sequence of quantum gates and measurements. You submit it as a "quantum task" specifying which device to target and how many shots (repeated runs) you want.
Braket queues the task, routes it to the right device, and returns results to your S3 bucket. For hybrid algorithms that alternate between classical and quantum steps, Hybrid Jobs coordinates both sides and gives you priority access to the QPU while your job is running.
A Few Concepts Worth Understanding
Quantum computing introduces vocabulary that can feel opaque at first. The course explains these well, but here's the short version:
- Quantum task: the atomic unit of work you send to a device. It's your circuit plus instructions on how many times to run it.
- Shots: because quantum results are probabilistic, you run the same circuit many times (10 to 100,000) and look at the distribution of outcomes. Each run is a "shot."
- Hybrid Jobs: for algorithms that alternate between classical computation and quantum computation (which is how most practical quantum algorithms work today), Braket coordinates both resources and gives your job priority QPU access.
That last one, Hybrid Jobs, is where things get interesting. Today's quantum computers aren't powerful enough to run entire algorithms end to end. The practical approach is to let a classical computer handle the heavy lifting and use the QPU as a specialised coprocessor for the parts where quantum mechanics provides an advantage. It's conceptually similar to how GPUs accelerate machine learning training.
Who Should Care
The course highlights three main use cases:
Researchers exploring quantum algorithms for finance, chemistry, logistics, or materials science. Braket lets you iterate quickly across multiple hardware types without vendor lock-in.
Teams evaluating quantum readiness for their business. You can benchmark quantum approaches against classical ones without committing to a single hardware vendor or building your own infrastructure.
Open-source developers building tools for the quantum community. Braket integrates with PennyLane (quantum machine learning), Qiskit (IBM's popular SDK), and supports Julia and OpenQASM in addition to Python.
What It Costs
Braket uses pay-as-you-go pricing. The Free Tier gives you 1 hour of simulator time per month, enough to work through the course demos. Beyond that:
- Simulators bill per millisecond of runtime
- QPUs charge a per-task fee plus a per-shot fee (varies by device)
- Notebooks are standard SageMaker notebook instances billed at EC2 rates
- Storage is standard S3 pricing for task results
The course makes a practical point here: always test on simulators first. QPU time is the expensive part, and you don't want to burn shots debugging a circuit that has a typo in it.
Getting Started (Literally)
The course includes four hands-on demos:
- Setting up Braket in the AWS console,
- Running a circuit on the SV1 state vector simulator (Free Tier eligible),
- Running the same circuit on a real IonQ QPU (this one costs money, reducible by lowering shots),
- Deleting resources used by Amazon Braket.
If you just want to experiment without any AWS charges, install the SDK locally (
pip install amazon-braket-sdk) and use the built-in local simulator. No AWS account needed for that.The Bigger Picture
AWS isn't just offering access to other companies' quantum hardware. They're investing in the technology itself through the Center for Quantum Computing (building fault-tolerant hardware), the Center for Quantum Networking (quantum communication research), and the Quantum Solutions Lab (a consulting programme that helps customers find quantum applications in their domain).
Braket is the customer-facing piece of that strategy. It's where you go to start writing code, running experiments, and building intuition for how quantum computing works in practice.
My Takeaway
Quantum computing is still early. Nobody is replacing their classical infrastructure with QPUs tomorrow. But the window for "this is only for physicists" has closed. If you can write Python and you're curious about what quantum can do for problems in optimisation, simulation, or machine learning, the tools are here, the barrier is low, and the cost of experimenting is a few dollars in simulator time.
The Amazon Braket Getting Started course is a solid first step. It won't make you a quantum physicist. It will show you that you don't need to be one.
Ready to try it? Start with the Amazon Braket Getting Started page, or explore the example notebooks on GitHub .
Series: Quantum on AWS, Without a PhD (2 articles)
- 1Your First Step Into Quantum Computing Doesn't Require a Physics PhD This article
Enjoyed reading this content? Let the author know!
Your likes, comments, shares, and saves help creators reach more builders.
Loading recommendations
Loading article