Executive brief
Most career guidance about medical imaging in Isère describes job titles. This dossier describes constraints, and derives the competences from them. Four dated facts do the work: a plant of 8,100 m² with 2,000 m² of cleanrooms building roughly 15,000 medical devices a year; a deliberate premium position taken against Asian price competition; a spectral-imaging programme, ImaSpiiR-X, supported by France 2030 with the CEA and the Lyon university hospital; and an accelerator constituted as a cooperative on 12 July 2024 whose declared service is pre-series assembly, certification and supply-chain construction. Together they specify seven competences. We also state the eight things we refuse to assert — starting with salaries, which no source we opened publishes for this site.
I. Why the constraints, not the vacancies
A job advert tells you what an employer wanted to write last quarter. An industrial constraint tells you what the employer will still need in five years. So the method here is deliberately inverted: we take facts the company and its regional ecosystem have put on the record with dates, and ask what work those facts make unavoidable.
The founding facts are these. Trixell SAS, Moirans (38430), SIREN 411 195 043, was incorporated on 26 February 1997 under NAF 2660Z — irradiation, electromedical and electrotherapeutic equipment (French business register, consulted 3 September 2026). It is a joint venture between Thales (51%), Philips Healthcare (24.5%) and Siemens Healthineers (24.5%), supplying large-format digital detectors for general radiography, fluoroscopy and cardiovascular imaging from a dedicated 8,100 m² plant including 2,000 m² of clean rooms (Trixell, about and shareholders pages, undated, consulted 3 September 2026). Its managing director since January 2024, Kaïs Mnif, describes mastery of the caesium iodide scintillator and amorphous silicon, roughly 15,000 devices a year, a refusal of the volume contest in favour of premium, and reported revenue of €200 million in 2024 with a €300 million ambition within three years (Les Affiches de Grenoble et du Dauphiné, 16 January 2025).
II. Seven competences the facts require
1. Cleanroom process discipline (2,000 m² of it). A cleanroom is not a room; it is a set of behaviours enforced by procedure — gowning, particle discipline, material entry rules, deviation reporting. The site's own description of a dedicated plant with 2,000 m² of clean rooms is the whole argument: a candidate who has never worked inside a controlled environment is not a marginally weaker hire, they are an untrained one. This is also the most accessible entry point on this list, because it is taught by employers rather than assumed.
2. Detector physics as applied craft. Scintillator plus amorphous-silicon readout is a specific chain with specific failure modes: conversion efficiency, lag and ghosting, pixel defect management, dose–image-quality trade-offs. The company frames its value precisely there — reducing radiation dose while optimising the image through exceptional resolution (Les Affiches, 16 January 2025). Reading that sentence as marketing is the mistake; it is a specification of what the engineering staff are measured on.
3. Reliability engineering, argued from consequence. The managing director's justification for extreme reliability is blunt: an imaging failure during a surgical procedure can be fatal to the patient (Les Affiches, 16 January 2025). Operationally that means accelerated life testing, failure-mode analysis, field-return forensics and statistical process control. In a premium strategy, reliability is not a department. It is the product's price justification.
4. Medical-device regulatory competence. The company's NAF classification is medical equipment manufacture, and AXEL's declared services name certification and market access explicitly (Auvergne-Rhône-Alpes Entreprises, published 9 September 2024, updated 22 September 2025). Design-history files, risk management, clinical-evaluation support and technical documentation are therefore permanent work, not a project phase. This competence travels: it is the same discipline anywhere in European medtech.
5. Industrialisation and pre-series engineering. AXEL exists to assemble industrial pre-series on pilot lines, build supply chains and master manufacturing cycles (same source). That is the named scarce competence of the region, stated by the institutions that built an accelerator around its scarcity. A candidate who can take a working prototype to a repeatable, certifiable pre-series is addressing a shortage that a cooperative was incorporated to relieve.
6. Spectral and computational imaging. ImaSpiiR-X is described as spectral imaging — colour X-ray images — supported by France 2030, with partners including the CEA and the Lyon university hospital (Les Affiches, 16 January 2025). The earlier Nemoxis project, led by Thales with partners including CEA-Leti, combined new-generation carbon-nanotube X-ray sources with image processing and artificial intelligence (Présences Grenoble, 29 March 2022). The competence implied is hybrid: detector physics plus algorithmic reconstruction. Pure software profiles and pure hardware profiles both under-fit it.
7. Cross-sector transfer of qualification. The company states that its X-rays now also explore industry: automotive batteries, airport security, agri-food inspection (Les Affiches, 16 January 2025). Each destination has its own standards, customers and duty cycles. The transferable asset is the qualification discipline, not the sensor — so the valuable person is the one who can restate a medical qualification argument in an industrial customer's vocabulary.
III. Where these competences are actually built
The regional route into this work is visible in AXEL's own founding list, which includes Université Grenoble Alpes, Grenoble INP and CHU Grenoble Alpes alongside industry (Thales, Fortil Group, Sherpa Engineering) and the CGT, with Grenoble Métropole, Le Grésivaudan and Pays Voironnais expected to join (Auvergne-Rhône-Alpes Entreprises, published 9 September 2024, updated 22 September 2025). We name those institutions because they are on the record as shareholders of the accelerator, not because we have measured their graduates' outcomes — which we have not.
Two structural notes for anyone planning a path. First, the accelerator's seat is the shared Thales–Trixell–Lynred site at Centr'Alp (ESSOR Isère, 4 December 2024): three imaging employers on one perimeter, which materially changes the risk of specialising in detector work at this location. Second, AXEL's director spent his career inside Trixell as a programme manager and head of partnerships and funding (same source) — evidence that programme and partnership roles, not only bench engineering, are career-bearing in this cluster.
IV. The scale question, answered honestly
How many people work there? Two dated figures exist and they do not agree. In March 2022 the Moirans industrial site was reported as employing 520 people — 120 for Thales and 400 for Trixell (Présences Grenoble, 29 March 2022). In January 2025 the company profile refers to 320 collaborators (Les Affiches, 16 January 2025). We publish both with their dates and decline to reconcile them, because no source we opened explains whether the perimeters differ or the workforce changed. A widely visible third figure — a platform profile claiming 130 employees and a revenue figure in the billions of US dollars (LinkedIn company page, consulted 3 September 2026) — contradicts the company's own reported revenue and is not used.
For a candidate this ambiguity is itself information: the site is in the low hundreds, not the thousands. Hiring is therefore episodic and role-specific, and the useful strategy is coverage of an adjacent employer set — the shared Centr'Alp perimeter, the accelerator's client start-ups, the broader imaging supply chain — rather than a single-employer plan.
V. One forward-looking number, carefully handled
In March 2022, Thales's radiology director stated that Nemoxis production should start in 2024–2025 and would require recruiting around one hundred people (Présences Grenoble, 29 March 2022). That is an intention recorded on a date, not an outcome. No source we opened states how many posts were created, whether the schedule held, or how the roles were distributed between Thales and Trixell. Anyone using that figure to plan a career move should treat it as a signal of direction — equipment integration rather than component supply — and not as an open requisition.
VI. Three honest candidate profiles
The cleanroom-capable technician. Entry is realistic through controlled-environment experience in any precision industry — semiconductors, photonics, sterile manufacturing — plus documented procedural rigour. The differentiator is deviation discipline: being the person whose records survive an audit. The regional density of cleanroom employers means this competence has more than one door.
The imaging engineer who can also argue. Detector physics or signal processing, plus the ability to write the qualification argument that convinces a regulator or an OEM customer. In a premium strategy explicitly opposed to volume competition (Les Affiches, 16 January 2025), the technical case for reliability is the commercial case. Engineers who can write it are structurally scarce.
The industrialisation generalist. Someone who has taken a device from prototype to certified pre-series at least once. AXEL was incorporated with €280,000 of capital, one employee at 4 December 2024, and a stated intention to identify the most-requested competences before hiring experts (ESSOR Isère, 4 December 2024). That is an unusually legible statement that this profile is being sought — and an equally clear signal that the organisation is small, so the opportunity is early-stage rather than salaried certainty.
VII. What we refuse to assert
Eight things are absent from this dossier on purpose: salary ranges or bonus practice at Moirans; the number of open positions today; hiring volumes by function; the site's current headcount as a single number; any success or conversion rate for AXEL's prospective start-ups; named training programmes we have not verified as feeding this employer; the ImaSpiiR-X or France 2030 funding amount attributable to this company; and any quantified skills shortage specific to Trixell. The market-standard framing in section I is the company's own description of its technology's position, reported as such, not a certified market share.
What is left is a defensible reading of a small, deep, part-owned-by-its-customers manufacturer that has chosen to be expensive and excellent, and of a cooperative built beside it to lend that excellence out. If you want a career here, the shortest honest route is not to learn a product. It is to become the person who can prove a device is fit to be sold.
