Engineering & Talent Solutions

Fab programmes hire in waves, and the waves want different things. Construction and tool install need a large, time boxed cohort in place before the schedule allows for it, sourced fast and stood down when the phase closes. The same programme needs a much smaller permanent process and equipment team that will still be there in three years, hired against a production organisation chart rather than a ramp date. Running both through one pipeline at one pace fails in both directions: the ramp cohort arrives after the tools do, and the permanent roles get filled with whoever happened to be available. TSD separates the two and sources each on its own timeline.

A rollout never concentrates its workforce in one place. Crews work dozens of sites at once, and the headcount a given market needs rises and falls week by week as the build front moves through it. The hiring problem is therefore one of mobilisation and demobilisation: teams have to be assembled where the next tranche of sites sits, then released rather than carried through the gap until the following phase is funded. Field roles add a second condition. Climbing tickets, working at height certification and rigging competence decide who is allowed onto a mast at all, and a lapsed certificate that nobody checked is discovered at the tower base on the morning it matters. TSD verifies competence before deployment, not on arrival.

Commissioning is a sequence of gates, and no level begins until the one beneath it has been signed off. That shapes the hiring more than the build does: the commissioning team has to be complete before level 4 opens rather than assembled during it, because an integrated systems test standing one engineer short moves every date behind it. Handover is a second and quite different problem. The people who prove a facility are not automatically the people who should run it, and critical facility operations wants technicians and shift engineers who will still be on the roster in five years, working live plant under permit. TSD treats the build phase and the operations phase as two hiring exercises on two timelines.

A ramp and a steady line reward different people. New product introduction wants engineers who are comfortable while the process is still moving, who can run a pilot build, absorb a design change and hold a schedule being rewritten around them. Steady state rewards the opposite temperament: somebody who will spend three years taking seconds out of a cycle time. Hiring one profile for both is how a plant ramps well and then drifts. The digitalisation layer is harder again, because an engineer genuinely fluent in both plant floor control systems and enterprise platforms is thin on the ground in any single labour market, and most applicants are strong on one side and merely conversant on the other.

A turnaround inverts the usual shape of a workforce. An operating asset runs for years on a small permanent core, then needs many times that number across a few intensely compressed days, every one of those people competent on the first morning because a shutdown offers no room to bring anyone up to speed once it has begun. The peak therefore cannot be filled late, and it cannot be filled approximately. Rotations set the terms on top of that. Offshore work is bounded by bed space, medical clearance and the travel that gets people out and back; onshore rotations still govern how long anyone stays. Safety certification gates all of it, and TSD checks it against the issuing register rather than the CV.

Validation sets an unusual hiring bar, because the deliverable is an approved document an inspector will read years later. An engineer who has done the equivalent task in an unregulated plant has not done this task. Writing and executing a protocol, handling a deviation properly and closing it so it survives an audit is a specific discipline, and adjacent experience does not transfer as cleanly as a CV implies. That narrows the field sharply and makes screening most of the work. The question is not whether somebody has worked in manufacturing, but whether they have worked inside a validated state and can show what they produced there. TSD screens for evidence of regulated work rather than proximity to it.

Technology skills go stale faster than a headcount plan can be approved. A migration wave wants cloud and platform specialists for eighteen months, while the same estate wants a service management and infrastructure team with no reason to change size at all. Filling both permanently overstaffs the organisation the moment the programme lands. Filling both on contract leaves nobody who remembers why anything was built the way it was, which is the more expensive mistake and the slower one to notice. Most engagements are therefore blended, and the real difficulty is deciding which roles sit on which side of that line before funding is committed rather than afterwards. TSD works that split with the client and sources each side to suit.

Project work is funded against milestones, and a milestone schedule pays no attention to a fiscal year. Design engineers are wanted heavily through detailed design and then hardly at all once drawings are issued; site engineering and supervision are wanted from mobilisation to handover and not a month before. An annual headcount plan cannot express that shape, so the workforce either sits idle between phases or arrives after the phase it was meant to carry has already started. Award dates move as well, which means the shape has to be agreed while the tender is live and triggered the day it lands. TSD builds against the milestone schedule and holds shortlists ready for an award date nobody can fix yet.

These are functional roles, and hiring them as generic functional roles is precisely what goes wrong. A finance manager who has only ever closed a monthly ledger will struggle with cost to complete on a contract running four years. An HR partner who thinks in job titles will not register that a lapsed certificate has just removed somebody from a live site this morning. What is wanted is functional depth plus enough fluency in how a project actually runs to be credible to the engineers who depend on it, and that pairing is under supplied because most functional careers are built on one side of it or the other. TSD screens for both and treats the second as a requirement.