Nuclear fission is the licensed craft of designing, fuelling, operating and emptying thermal-spectrum and fast-spectrum nuclear reactors that convert fission heat to electricity. The work runs through reactor physics, nuclear fuels, reactor materials, nuclear structural materials, nuclear safety, nuclear waste management and nuclear power plant operation. At the end of 2025 the IAEA counted 413 operational units with 377.1 GW(e) of capacity and 2635.3 TWh of generation, with 72 reactors totaling 76.9 GW(e) still under construction . Two thirds of operational capacity has already passed 30 years of service, so the same specialists keep ageing cores inside their analyses of record and staff new units whose fuel, lattice and licensing basis do not interchange with the fleet they came from .
Challenges in Nuclear Fission Recruiting
Subsequent license windows for ageing nuclear reactors
Most of the people this craft still needs are not designing first concrete. About 66.5% of global operational capacity, 293 units, has already run more than 30 years, and 190 units, 41% of installed capacity, have passed 40 years, while the IAEA's 2026 high case still more than triples world capacity to 1284 GW(e) by 2060 only if licence renewals, uprates and plausible construction all hold . In the United States the NRC has defined subsequent license renewal as the period of extended operation from 60 years to 80 years, and a growing list of pressurized-water and boiling-water plants already holds that paper . The hiring implication is specific. A greenfield core designer who has never revalidated a time-limited ageing analysis is not interchangeable with the engineer who must show that neutron fluence, fatigue usage and environmental qualification still bound the updated final safety analysis report. Programmes that staff "nuclear engineers" against a 2060 capacity chart without saying whether the seat is a 40-year-old PWR, a new VVER-TOI, or a PHWR reload will shortlist people who cannot sign the chapter the regulator will actually read.
Nuclear fuels that split LEU oxide from HALEU qualification
The existing U.S. fleet is licensed on low-enriched uranium, fuel enriched to no more than 5 weight percent U-235, almost always as uranium oxide pellets in cylindrical zircaloy or ZIRLO cladding . Most advanced reactors under development in the United States instead require high-assay low-enriched uranium, greater than 5 and less than 20 weight percent U-235, to reach smaller cores, longer cycles and higher power density, and DOE states that HALEU is not currently available from domestic commercial suppliers . The Department estimates domestic demand could reach 50 metric tons per year by 2035; the Centrus demonstration cascade at Piketon has produced more than 100 kilograms and is expected to ramp toward 900 kilograms per year, which DOE itself describes as insufficient for long-term requirements . That gap is a fuel-qualification problem, not a generic materials shortage. A reload engineer who has specified gadolinia burnable absorbers and grid-to-rod fretting on a 17x17 PWR assembly has not qualified TRISO particles, metallic sodium-bonded pins, or HALEU deconversion from UF6. Reactor materials for cladding and pellet-cladding interaction sit in that same split: ZIRLO experience does not travel to SiC composite cladding or to stainless systems in a fast spectrum. Hiring that treats "fuels engineer" as one market will put a LEU oxide specialist against a HALEU irradiation campaign and call the mismatch a recruiting delay.
Reactor physics locked to lattice, coolant and burnup methods
The title reactor physicist is shared across methods that do not share a licensed toolchain. For light-water cores, SCALE/TRITON is used to generate burnup-dependent, few-group homogenized cross sections, discontinuity factors, pin powers and kinetic parameters for nodal simulators such as PARCS, with branch cases in fuel temperature and moderator density that the core simulator later interpolates . That lattice-to-core chain is not a Monte Carlo shielding calculation, and it is not interchangeable across coolants. A PWR reload is organized around soluble boron letdown and assembly shuffling; a BWR core is void-fraction dominated; a PHWR lattice is D2O-moderated natural uranium with on-power refuelling. The IAEA's 2025 operating record still distinguishes those types in performance: BWRs and PWRs have been the best-performing reactors over the past decade, with load factors of 90.4% and 81.3% respectively, which is another way of saying the operating envelope, and therefore the physics that supports it, is plant-class specific . A CV that lists MCNP, SCALE or "core design" without naming the lattice, the few-group library, the nodal code and the cycle that went to the safety committee is advertising a toolkit. The question that matters is which depletion sequence and which simulator produced the analysis of record.
Nuclear structural materials under vessel fluence limits
The vessel is the component a utility cannot swap, and the people who keep it inside its limits are not the same people who specify fuel cladding. IAEA SSG-48 treats time-limited ageing analyses as safety analyses that use time-limited assumptions: neutron fluence as the time-dependent variable, neutron embrittlement of the vessel material as the associated ageing effect, compared against a regulatory limit before continued service is accepted . Surveillance capsules, ductile-to-brittle transition temperature shift, and pressurized thermal shock are the evidence, not a general "materials" narrative. Nuclear structural materials in this sense are the pressure boundary, the internals swelling case, and the containment liner; reactor materials in the fuel and cladding sense degrade on a reload cycle and are designed to be replaced. Subsequent license renewal in the U.S. is built on the Generic Aging Lessons Learned for Subsequent License Renewal report and on ageing-management programmes that must remain consistent with that guidance through 80 years . A candidate who has characterized irradiated cladding or spacer grids has useful reactor-materials depth and still may never have owned a vessel integrity chapter. Mix those seats and the plant discovers the gap when the fluence projection, not the pellet, is what the regulator asks to see.
Nuclear waste management after the generating unit stops
Spent fuel does not follow the operating licence off site. On 4 August 2026 Finland's Radiation and Nuclear Safety Authority stated that Posiva's encapsulation and final disposal facility at Olkiluoto meets the safety requirements for an operating licence, the first time a geological repository for spent nuclear fuel has reached that phase anywhere . The high-level fuel is sealed in iron and copper canisters in an above-ground encapsulation plant, then placed 400 to 450 metres down, with bentonite and bedrock as complementary release barriers; STUK and external experts have spent more than 250 person-years on the project since 2000, and even after a government licence STUK must still confirm readiness before emplacement starts . That is a different profession from wet-pool cooling, dry-cask loading, or vitrified high-level waste from a reprocessing line. Nuclear waste management hiring fails when a decommissioning manager, a cask-licensing engineer and a repository safety-case author are treated as one pipeline. The Olkiluoto file is a reminder of the timescale: the generating unit's outage is measured in weeks; the waste package is argued over decades and then over geological time.
Outage ownership inside nuclear power plant operation
Nuclear power plant operation is scheduled in outage windows, not in headcount plans. Global load factor in 2025 was 81%, with more than a third of net electrical capacity above 90%, which is possible only if planned unavailability is held to a critical path the grid already booked . The people who hold that path are shift technical advisors, system engineers who own a train through the outage, fuel-handling supervisors, and outage managers who can freeze scope when a finding appears on the vessel or steam generator. A commissioning engineer from a new-build site, a simulator instructor, and a control-room operator on a different NSSS vendor do not share that ownership even when the job title is "operations." The cost of a miss is concrete: a day of lost generation on a gigawatt-class unit, contractor surge into a window that cannot move, and a restart that waits on the same small group who can sign the mode-change. Hiring that screens for a licence without asking which outage, which NSSS, and which system the candidate owned will fill the roster and still slip the window.
Nuclear safety evidence the analysis of record actually owns
Nuclear safety in this craft is not a culture statement. IAEA SSR-2/1 (Rev. 1) requires that design incorporate defence in depth, with independent levels so that a failure is detected and compensated, applied in full power, low power and shutdown states . In the U.S. licensing file that requirement becomes paper a named engineer can point to. 10 CFR 50.34 requires a final safety analysis report that describes the facility, the design bases, the limits on operation, and a safety analysis of structures, systems and components; ECCS cooling performance after a postulated loss-of-coolant accident must be evaluated under 10 CFR 50.46 . For light-water reactors fueled with uranium oxide in zircaloy or ZIRLO, the calculated peak cladding temperature shall not exceed 2200 °F, cladding oxidation and hydrogen generation are capped, the core must remain in a coolable geometry, and long-term decay heat removal must be shown; a change or error that moves peak cladding temperature by more than 50 °F is significant and reportable . Verification therefore has a short list. Which FSAR chapter did the candidate own. Which ECCS evaluation model, realistic or Appendix K, produced the limiting break. What independent verification challenged the result, and what changed. The cost of getting that assessment wrong is not a weak quarter: it is a reanalysis the plant cannot start up without, an operating restriction while the model is rebuilt, and senior reviewers spending months on a shortlist that never held a licensing number.
References
- Nuclear Power Reactors in the World, 2026 Edition (RDS-2/46) — International Atomic Energy Agency (IAEA). (accessed 2026-09-27)
- Energy, Electricity and Nuclear Power Estimates for the Period up to 2060 (RDS-1/46) — International Atomic Energy Agency (IAEA). (accessed 2026-09-27)
- Status of Subsequent License Renewal Applications — U.S. Nuclear Regulatory Commission (NRC). (accessed 2026-09-27)
- HALEU Frequently Asked Questions — U.S. Department of Energy (DOE). (accessed 2026-09-27)
- SCALE/TRITON Primer: A Primer for Light Water Reactor Lattice Physics Calculations (NUREG/CR-7041) — U.S. Nuclear Regulatory Commission / Oak Ridge National Laboratory. (accessed 2026-09-27)
- Safety of Nuclear Power Plants: Design, Specific Safety Requirements SSR-2/1 (Rev. 1) — International Atomic Energy Agency (IAEA). (accessed 2026-09-27)
- The Radiation and Nuclear Safety Authority supports the licence for a spent nuclear fuel final disposal facility — Radiation and Nuclear Safety Authority of Finland (STUK). (accessed 2026-09-27)
- Ageing Management and Development of a Programme for Long Term Operation of Nuclear Power Plants (SSG-48) — International Atomic Energy Agency (IAEA). (accessed 2026-09-27)
- 10 CFR 50.34 and 50.46 — Contents of applications; ECCS acceptance criteria — U.S. Nuclear Regulatory Commission (NRC). (accessed 2026-09-27)