Quantum Machine Learning Engineer

A quantum machine learning engineer sits at the intersection of classical ML and quantum computing. The job is to build hybrid pipelines where a parameterised quantum circuit becomes a trainable layer inside an otherwise classical model, then to test honestly whether that circuit buys you anything a classical network would not. It is a research-leaning engineering role: equal parts PyTorch fluency, quantum-circuit design, and disciplined benchmarking.

Est. base salary (US) $115k - $240k+
Focus Hybrid ML, variational circuits
Best for ML engineers and data scientists

A typical day

Most of the work happens in a notebook or training script. A QML engineer spends the morning iterating on a feature map and ansatz in PennyLane, watching gradients to see whether the model is actually learning or stuck on a barren plateau. The afternoon is often spent on the classical side: wiring the quantum layer into a PyTorch model, tuning the optimiser, and comparing results against a strong classical baseline. Periodically experiments graduate from a noiseless simulator to a real QPU over the cloud, where shot noise and hardware error become the dominant concern. Time is also spent reading recent papers, since the field moves quickly and many techniques are only a year or two old.

Core responsibilities

  • Build hybrid quantum-classical pipelines that combine classical deep learning with parameterised quantum circuits.
  • Design data-encoding (feature map) schemes that load classical data into quantum states without destroying the information you need.
  • Implement and tune variational models: variational classifiers, quantum kernels, and generative circuits.
  • Train circuits with gradient-based optimisers, diagnosing problems like barren plateaus and shot noise.
  • Benchmark quantum models honestly against strong classical baselines to establish whether any advantage is real.
  • Run experiments on simulators first, then on real QPUs over the cloud, accounting for hardware noise.
  • Translate domain problems (chemistry, finance, vision) into formulations a near-term device can attempt.
  • Communicate results and limitations clearly to research leads and non-specialist stakeholders.

Skills

Required

  • Classical machine learning (PyTorch or TensorFlow)
  • Python (NumPy, autodiff, optimisation)
  • PennyLane
  • Variational circuits and ansatz design
  • Quantum kernels and feature maps
  • Data encoding strategies
  • Gradient methods and optimisation
  • Linear algebra and probability

Nice to have

  • VQE and QAOA
  • TensorFlow Quantum
  • Qiskit Machine Learning
  • JAX and GPU-accelerated simulation
  • Quantum chemistry or finance domain knowledge
  • Barren plateau mitigation
  • Error mitigation on real hardware
  • Published research or preprints

Tools of the trade

  • PennyLane

    The dominant differentiable QML framework, with autodiff that plugs straight into PyTorch, TensorFlow, and JAX.

    Learn more →
  • PyTorch / TensorFlow

    Classical deep-learning backbones for the hybrid half of the model, plus the optimiser ecosystem.

  • TensorFlow Quantum

    Google framework for integrating parameterised quantum circuits into TensorFlow training loops.

    Learn more →
  • Qiskit + simulators

    Circuit construction, transpilation, and noisy simulation before running on real IBM hardware.

    Learn more →
  • Cloud QPUs

    Amazon Braket and Azure Quantum give access to multiple hardware backends for real-device experiments.

    Learn more →

Salary by seniority

Approximate US base-salary ranges for 2026. Total compensation at well-funded labs adds meaningful equity and bonus on top. Bay Area, NYC, and Boston typically add 20-40%.

LevelBase rangeWhat changes
Junior (0-3 yr) $115k - $145k Hybrid pipelines, PennyLane / PyTorch, simulator-based experiments.
Mid (3-7 yr) $145k - $185k Owns model design, variational ansatz choice, and benchmarking against classical baselines.
Senior / Staff (7 yr+) $185k - $240k+ Sets QML research direction, publishes, and translates business problems into quantum-amenable formulations.

See the full quantum computing salary guide for geographic breakdowns and the skills that command a premium.

Demand and outlook

Quantum machine learning is one of the faster-growing sub-fields, driven by interest from finance, pharma, and materials companies looking for early use cases. Demand is real but more selective than for general quantum software roles, because employers increasingly want engineers who can prove or disprove advantage rather than simply run demos. The strongest candidates pair genuine classical ML depth with quantum-circuit fluency, and that combination remains scarce. Expect the bar to favour people who understand both barren plateaus and gradient descent equally well.

Who hires for this role

  • Xanadu
  • IBM
  • Google Quantum AI
  • AWS
  • Microsoft
  • Quantinuum
  • IonQ
  • Multiverse Computing
  • Banks and hedge funds
  • Pharma and materials R&D labs

Browse current openings on the quantum jobs board, and see how this role fits alongside others in the careers overview.

How to become a quantum ML engineer

The most direct path is to build solid classical ML skills first, then layer quantum circuits on top using PennyLane. Our step-by-step roadmap walks through the full sequence, from foundations to landing the role.

Read the full guide: How to become a quantum ML engineer →