Building the team — roles open and the org chart

Why this team looks different from a software team

Building a 100-physical-qubit superconducting transmon machine in Co. Tipperary is not a software hire. It is a low-temperature physics, microwave engineering, FPGA control, and systems-software hire — all running in parallel against a twelve-month delivery window. The org chart we are putting in place reflects that. It is deliberately flat for the first eighteen months because most decisions cross at least three disciplines: a change in the qubit chip floorplan touches the wiring tree inside the dilution fridge, which touches the room-temperature control electronics, which touches the pulse-level compiler. There is no clean handoff. The people we hire need to be comfortable with that.

Below is a working picture of the open roles, what each person actually does day-to-day, and how the disciplines stack against the Q3 2026 → Q2 2027 milestones. If you are looking at quantum careers in Ireland and you want to know whether the work is real, this is the page to read.

The cryogenic and hardware stack

The machine sits inside a dilution refrigerator operating below 15 mK at the mixing chamber. Above that, there are five further temperature stages — 4 K, still plate, cold plate, and intermediate plates — and every signal line, every attenuator, every isolator and circulator has to be heat-budgeted at each stage. A 100-qubit system with readout multiplexing needs roughly 200–300 coaxial lines descending into the fridge. Each input line carries ~60 dB of attenuation distributed across stages to suppress thermal photons; each output line carries TWPA (travelling-wave parametric amplifier) and HEMT amplification stages on the way out. Get the attenuation budget wrong and qubit T₁ times collapse.

Roles open in this layer:

• Cryogenic Systems Engineer — owns the fridge, the wiring tree, the magnetic and IR shielding, vibration isolation, and the helium plant interface. Hands-on experience with Bluefors- or Oxford-class systems is the baseline.
• Microwave Engineer — designs and characterises the room-temperature signal chain: AWG outputs, IQ mixers, LO distribution, readout demodulation. Comfortable with VNA work, S-parameter modelling, and intermodulation budgets in the 4–8 GHz band.
• Quantum Hardware Engineer (qubit packaging) — owns the chip carrier, wirebonding, TSV interposers where used, and the magnetic shielding can. This is where heavy-hex topology meets physical reality: you cannot route a heavy-hex coupling graph cleanly without a multi-layer interposer once you are past about thirty qubits.

The control and calibration layer

A transmon is calibrated, not configured. Every qubit has a slightly different frequency, anharmonicity, T₁, T₂*, and readout dispersive shift. The calibration graph — the dependency tree of measurements that has to be re-run when a parameter drifts — is a nontrivial DAG with hundreds of nodes for a 100-qubit device. Automating that DAG, scheduling it, and keeping the machine in spec without human intervention is one of the hardest unsolved problems in the field.

Roles open in this layer:

• Quantum Control Engineer — pulse-level work in the FPGA control stack. Writing DRAG pulses, cross-resonance gates, parametric two-qubit gates depending on the coupler architecture, and characterising leakage out of the computational subspace into the |2⟩ state.
• Calibration Software Engineer — owns the automated calibration DAG. Python, with hard real-time constraints pushed down to the FPGA layer. Familiarity with tools in the Qiskit Pulse / qiskit-experiments idiom, or equivalent in-house frameworks, is useful but not required if the systems instinct is there.
• Quantum Characterisation Scientist — runs randomised benchmarking, gate set tomography, cycle benchmarking, and crosstalk characterisation. Publishes the fidelity numbers we will report against.

The error-correction and compiler track

One hundred physical qubits is not one hundred logical qubits. The roadmap to surface-code logical qubits requires distance-3 patches at minimum (17 physical qubits per logical, plus ancillas) and realistically distance-5 (49+) before a single logical qubit beats its constituent physical qubits on any useful workload. We are honest about that. The team building toward it now is small but the work starts on day one because the control stack has to be designed with mid-circuit measurement, fast feed-forward, and decoder latency in mind from the start.

• Quantum Error Correction Engineer — surface code, decoder implementation (Union-Find or matching-based), syndrome extraction circuits, and the latency budget between measurement and Pauli-frame update.
• Compiler Engineer — OpenQASM 3 front-end, transpilation against the heavy-hex coupling map, SWAP routing, scheduling under T₁/T₂ constraints, and the lowering path to the pulse layer. Work touches Qiskit, PennyLane and Cirq because customer code arrives in all three.

The applications team — chemistry first

The reason the priority cohort is climate workloads is that the near-term advantage for noisy intermediate-scale machines lives in variational chemistry, not in factoring. VQE and its successors (ADAPT-VQE, contextual subspace methods) on active spaces of 20–50 spin-orbitals are within reach of a well-calibrated 100-qubit device, and that bracket covers transition-metal catalysts relevant to direct air capture, MOF linker design, and the redox chemistry inside next-generation flow batteries.

• Quantum Chemistry Scientist — active-space selection, Hamiltonian construction in second quantisation, fermion-to-qubit mappings (Jordan-Wigner, Bravyi-Kitaev, parity), and ansatz design. Background in CASSCF/CASPT2 is more useful here than a pure quantum-computing background.
• Climate Applications Lead — translates between research partners working on carbon capture, photovoltaic absorbers, and grid optimisation, and the chemistry/optimisation teams. The output of this role feeds directly into IMPT's offset-stack supplier evaluation, which is the main reason the integration exists.

• Research collaboration or early access

  Direct with Michael. No charge for the call.

  Book a research call →