Magnetoresistive-Memory Reliability Recovery Leader — Aerospace Compute

The mandate

The emerging-memory reliability head departed after combined temperature and radiation testing revealed rare write failures, retention loss and read disturbance outside distributions used for aerospace qualification. Device teams attribute tails to test stress, while controller teams propose stronger correction that may conceal progressive margin loss. The interim leader must recover a mechanism-based qualification position without allowing schedule or scarce radiation-beam time to narrow representative evidence.

Ten months provide the recovery window. The first six weeks secure samples, test histories and process splits and establish competing device, array and instrumentation hypotheses. Months two through six complete mechanism-stable electrical, thermal and radiation studies. Months seven through nine qualify screens, controller limits and production monitoring. The final month installs a permanent leader and transfers customer, foundry and residual-lifetime decisions.

Handover is achieved when write-error, retention and disturb tails meet approved mission limits across qualified lots; radiation responses are reproduced at two facilities or reconciled technically; and correction load remains within a transparent budget. The successor must chair one lot disposition, approve a stress-model revision and defend residual lifetime uncertainty to aerospace assurance. Excluded samples and uncorrectable tails must remain visible.

The leader may quarantine lots, approve stress matrices, allocate beam and chamber time, set screening thresholds, direct foundry containment and commit EUR 6 million within the authorised recovery budget. They decide product-reliability readiness inside approved aerospace requirements. Memory-cell redesign, mission classification, formal radiation assurance, customer contractual relief and long-term foundry awards remain with named technical, quality and commercial executives.

Excluded are redesigning the complete compute module, owning spacecraft or aircraft hazard analysis, acting as accredited radiation authority or replacing the memory controller platform. The mandate covers MRAM device-to-product lifetime evidence and sustainable volume controls. Patent conclusions, supplier liability and customer fleet action remain outside interim authority, supported by preserved samples and traceable results.

Why this seat is open

MRAM reliability tails arise from coupled magnetic, electrical, thermal and radiation effects that average correction rates can conceal. The leadership gap leaves device and controller groups debating attribution while qualification time continues to expire. Temporary executive authority can preserve scarce evidence, make conservative lot decisions and leave a permanent leader with repeatable mechanisms and transparent correction budgets.

What you will own

• Reconcile wafer lot, array location, magnetic state, pulse conditions, temperature history, radiation exposure and test-equipment genealogy.
• Separate stochastic write error, retention loss, read disturb, peripheral-circuit response, recovery and instrumentation artefacts.
• Design combined electrical, thermal and radiation studies with mechanism checks and representative aerospace duty cycles.
• Evaluate error correction, retry and scrubbing for margin visibility, latency, power and uncorrectable-tail consequences.
• Establish lot screens and monitors that predict mission risk without damaging cells or filtering legitimate extremes.
• Direct foundry, test-house and controller corrective actions through sample preservation, split evidence and qualification change control.
• Transfer lifetime models, radiation discrepancies, open lots, customer commitments and successor-led disposition evidence.

Candidate qualifications

• Led MRAM or emerging-memory reliability qualification for aerospace, defence, automotive or another high-consequence electronics market.
• Can evidence a rare write or retention tail obscured initially by average correction or screening results.
• Understands magnetic switching, write margins, retention, disturb, radiation effects, arrays and statistical reliability inference.
• Has governed scarce beam time and cross-facility correlation without overfitting conclusions to one test setup.
• Integrated device, controller, foundry and product-assurance evidence under an external qualification deadline.
• Developed a successor able to defend transparent tail risk rather than relying on a reassuring aggregate error rate.

Non-negotiables

• Will attend Brussels, Leuven, Grenoble and Dresden laboratory, beam and foundry sessions.
• Has personally dispositioned emerging-memory lots using combined thermal, electrical and radiation evidence.
• Accepts no authority over mission hazards, accredited assurance findings, customer fleet actions or foundry awards.
• Will disclose memory, foundry, radiation-laboratory, aerospace and controller-IP relationships before appointment.

1. 49 words maximum. Which MRAM tail remained hidden longest behind correction or aggregate reliability results?
2. 49 words maximum. How would you reconcile conflicting radiation findings from two qualified facilities?
3. 49 words maximum. What must the successor demonstrate before inheriting lot-disposition authority?