Structural Engineer Interview Questions

I start by establishing the load case before anything else: dead loads from the structure's own weight and finishes, imposed loads appropriate to the building's use, and environmental loads from wind, snow, or seismic action depending on the site. I build a simple load takedown by hand first, tracing the path from slab to beam to column to foundation, so I have an independent check before any model is built. Once I understand roughly what the structure needs to carry, I sketch two or three viable structural systems and compare them against the architectural intent, the span requirements, and the construction programme. I then build the detailed model, typically in ETABS or a similar package, size the primary members, and run the code checks for strength, deflection, and stability. I always cross-check a handful of critical members against my hand calculations before I trust the software output. Only once the scheme is stable and the numbers hold up do I move into detailing, connection design, and coordination with the other disciplines. Skipping the hand-calculation stage is how engineers end up trusting a model that has an input error buried in it.

The decision usually comes down to span, speed, fire performance, and the client's cost and sustainability priorities, and there is rarely a single right answer. Steel gives you long clear spans and a fast, dry erection programme, which matters when the client needs the building watertight quickly, but it needs fire protection in most occupancies and its embodied carbon per tonne is higher than timber. Concrete gives you inherent fire resistance and mass that helps with acoustic separation and thermal performance, and it is often the cheaper option for shorter spans, but the construction programme is slower because of curing time and it carries a significant embodied carbon cost from cement. Timber, particularly cross-laminated or glulam, has become a serious option for mid-rise buildings: it is fast to erect, has a strong carbon story, and performs predictably in fire when sized correctly for char rate, but span limitations and moisture management need careful thought. On a recent project I recommended a hybrid: a timber superstructure on a concrete podium, which gave the client the carbon reduction they wanted on the upper floors while keeping the ground floor commercial units in a material that handled the higher loading and fire strategy more simply.

Design conflicts are normal, and I have learned that the fastest way through one is to separate the architectural intent from the specific solution the architect has proposed. On one project the architect wanted an exposed steel transfer beam left visible in a double-height lobby, but the beam depth needed for the span made it visually heavier than the design intent allowed. Rather than simply telling them the beam had to be deeper, I asked what the intent behind the exposed steel actually was: it turned out the priority was a visible structural moment in the space, not that specific beam shape. That let me propose a Vierendeel truss across the same span, which kept the visual lightness they wanted while giving me the depth I needed structurally. I try to get into these conversations early, before a scheme is fixed in a way that makes changes expensive, and I bring options rather than just objections. Architects respond much better to "here are two ways to get what you want structurally" than to being told no.

My core toolkit is ETABS for building structures, particularly anything with lateral load considerations like wind or seismic, and SAP2000 when I need more flexibility for non-standard geometries or specialist structures like bridges or industrial frames. For steel connection design and detailing coordination I use Tekla Structures, which also gives me a clash-detection view against the architectural and services models. For quick sizing and sanity checks I still reach for spreadsheets with the relevant code equations built in, because a spreadsheet I have built and verified myself is often faster and more transparent than opening a full model for a single beam. I choose the tool based on what the structure actually needs: a simple braced frame does not need the same modelling effort as a transfer structure with irregular load paths. The tool is only as good as the engineer's understanding of what it is doing internally, so I never treat any of them as a black box, especially for anything involving dynamic or nonlinear behaviour.

During construction of a mezzanine extension, I was doing a routine site visit when I noticed the contractor had substituted a lighter gauge steel section for the specified beam on two bays, apparently to use up stock they already had rather than through any formal request. On paper the difference looked small, but when I ran the numbers that evening the substituted section failed the deflection check under full imposed load by a meaningful margin, and it was already installed with the floor slab poured on top. I stopped work on that section the same day, which was not a popular call with the contractor given the programme pressure they were under, and I informed the client and the principal designer immediately with a written record of what had happened and why it mattered. We assessed strengthening options and settled on bolting additional plates to the underside of the affected beams, verified by calculation and then load-tested before we allowed the area back into use. It added about a week to that part of the programme. I would rather absorb that delay and an uncomfortable conversation with the contractor than sign off a floor I was not confident in.

On a warehouse project the client's cost consultant challenged the wind bracing I had specified on the gable end, arguing it was over-designed compared with a similar building they had built elsewhere and that removing two bracing bays would save a meaningful amount on the steel package. I did not simply hold my ground on authority. I went back to the wind load calculation, checked the terrain category and the building's exposure against the actual site rather than an assumption, and confirmed the loading was correct for that specific location, which had a longer unobstructed fetch than the comparison building they were citing. I put together a short technical note explaining the difference in exposure between the two sites and what would happen to the bracing utilisation if we removed the bays they wanted cut, which showed the frame would exceed 100 percent utilisation under the design wind case. Once they could see the number rather than just my say-so, the pushback stopped. I also offered an alternative saving elsewhere in the frame that did not touch the lateral system, which meant they still got some of the cost reduction they were after.

I was running structural design on three projects at once for about two months last year: a school extension in detailed design, a residential scheme going through planning, and a warehouse in early concept. I kept a simple weekly priority list ranked by which project had the nearest hard deadline, since not all project work is equally urgent at any given moment, and I blocked specific days for deep design work rather than trying to context-switch hour to hour, which I have found kills calculation accuracy. I was explicit with each project lead about what response time they could expect from me that week, rather than letting them assume I was fully available and then disappointing them. The school extension had a planning deadline that could not move, so that got priority when the three projects actually collided. I also delegated the more routine connection design on the warehouse to a graduate engineer I was mentoring, with checks built in at defined points, which freed up my time for the more judgment-heavy items on the other two projects and gave them useful development experience.

For wind design I start by establishing the basic wind speed for the site from the relevant code, then adjust for terrain category, altitude, and the building's exposure, which for a mid-rise structure often means checking both the overall building response and localised pressures at corners and parapets where suction can be significantly higher than the average face pressure. I then calculate the overturning moment and base shear the lateral system needs to resist, and check that against the bracing, shear walls, or moment frame I have proposed, paying attention to torsional effects if the building's stiffness is not symmetric. For seismic design in a region where it governs, the starting point is the seismic zone and the ground type, which sets the design spectrum, then I calculate the base shear using either an equivalent static method for regular structures or a response spectrum analysis for anything with significant irregularity in plan or elevation. Ductility matters as much as strength in seismic design: I detail connections and reinforcement to allow the structure to deform and dissipate energy rather than fail suddenly, which is a different design philosophy from simply designing for a higher static load. I always check which load case, wind or seismic, actually governs for a given structure rather than assuming, because it varies by building height, mass, and location.

Buckling is a stability failure rather than a material strength failure: a slender member under compression can fail by suddenly deflecting sideways at a load well below the stress that would crush the material itself, because the geometry of the member becomes unstable before the material reaches its capacity. The classic example is a slender steel column, where the critical buckling load depends on the member's stiffness, its length, and crucially its effective length, which accounts for how the ends are restrained. A column pinned at both ends has a different effective length, and therefore a very different buckling capacity, than the same column fixed at both ends, even though the material and cross-section are identical. In design I check the slenderness ratio against the code limits and apply the appropriate reduction factor to the member's compressive resistance, and for anything unusually slender I look at whether intermediate restraints, bracing, or a stiffer section is the more efficient fix. Buckling also shows up locally, in thin-walled sections or plate elements, which is why I check local buckling of flanges and webs separately from overall member buckling, particularly in cold-formed or built-up steel sections where the plate thickness is small relative to its width.

I treat this as a deliberate part of my working week rather than something that happens by accident. I follow the update notices from the standards bodies relevant to my market directly, since code amendments can change a design assumption that was previously safe, and I make a point of reading the commentary behind a change, not just the clause itself, because understanding the reasoning helps me apply it correctly to situations the code did not explicitly anticipate. For new materials, I look for published test data and case studies rather than manufacturer marketing claims, and where possible I speak to engineers who have already used a material on a completed project to hear what went wrong as well as what went right. I have also built a habit of running a lessons-learned note after any project where I used something unfamiliar, whether that is a new connection type or a new analysis method, so the knowledge does not just live in my head. On timber engineering specifically, the guidance has moved quickly over the past few years as mass timber has become more common in mid-rise buildings, so I have taken specific continuing professional development courses on that rather than assuming my general structural training covered it.