Aarush Yusuf.

I design the piston that physically separates IPA and N₂O inside the propellant tank assembly of a reusable liquid rocket engine. It is a safety-critical part: if the seal between the two propellants fails, they meet somewhere they are never meant to meet.

The design starts as hand calculation and progresses into parametric CAD in Onshape, so the geometry stays driven by the numbers rather than by what happens to look right on screen.

The seal work is the core of it. I develop the O-ring groove sizing — squeeze factor, stretch, and volumetric gland fill — to hold across a 500–800 psi range. Getting gland fill right matters because an over-filled groove has nowhere to displace to under pressure, and an under-squeezed one leaks.

I founded and led TNO (The Named One) through the complete UKSEDS National Rocketry Championship lifecycle — PDR, CDR, Manufacturing Report, and Flight Readiness Review — to a live launch that reached 1,700 ft apogee with real-time telemetry streaming from onboard sensors to a ground station.

Leading ten people taught me something I did not expect to be the hard part. Full-team meetings were quietly failing: subteams sat through an hour of discussion that was 80% irrelevant to them and left no better aligned. I restructured it — took structures leadership myself, delegated avionics leadership to my vice-lead, and instituted weekly cross-team syncs instead. Alignment problems are usually structural, not effort problems.

On the engineering side I modelled the full airframe in Fusion 360 and verified every OpenRocket simulation before submission. When the avionics came in under the 300 g mass I had designed around, that was not a free win — the stability margin depended on that mass being where I assumed it was. I added ballast to the nosecone to restore the centre-of-gravity position the structural design was built on.

I have held two roles on Sheffield EcoMotorsport as the team builds toward the iLumen European Solar Challenge.

Since July 2026 I have been an Aerodynamics & Mechanical Design Engineer, manufacturing carbon fibre bodywork for SEM03 and contributing to early design work on SEM04 for the iLumen 2028 competition.

Before that, as Powertrain Engineer, I iterated the battery casing design across six versions in Fusion 360 — moving through L-bracket and fibreglass construction approaches — to carry a 7.5 kg pack. Each version went through Fusion static FEA against a 20 g load requirement, which is what validated the final 300 × 280 × 3 mm mild steel plate supported on six 12 mm-diameter rods.

I designed and scratch-built a 600 mm wingspan fixed-wing RC aircraft from component level. The motor/ESC/battery combination came out of datasheet analysis rather than guesswork — the selected setup draws 20.53 A, which sits at 68% of the battery's 30 A continuous ceiling, leaving real headroom instead of running the pack at its limit.

The first motor burned out. The cause was a prop-size mismatch: I was running a drone motor outside its thermal envelope, because a fixed-wing operating range asks for sustained load in a way a quadcopter's duty cycle never does. Diagnosed it, corrected the pairing.

Then it kept breaking in flight, repeatedly, in the same way. I worked through three iterative test cycles to find out why. Frame-by-frame flight video analysis put the centre of gravity at roughly 60% MAC against a 25–33% safe range — badly aft, which makes the aircraft violently pitch-unstable. I corrected it to 22% MAC.

At Haqqathon 2026 at Imperial College London I led a four-person multidisciplinary team to design and prototype a myoelectric below-elbow prosthetic arm, targeting amputees in conflict-affected and low-income communities.

The whole project was organised around one constraint: unit cost. We reached £72 per unit — around 95% cheaper than commercial alternatives. That number is the point of the project, not a footnote to it, because a prosthetic that the intended user cannot buy has not solved anything.

I designed the full hand assembly in Fusion 360: a modular five-finger mechanism in ABS with a parametric fit to stump geometry, and bolt-level replaceability throughout. That last decision matters more than it sounds — when a single finger fails on a conventional unit, the user replaces the whole arm. Here they replace the finger.

A 450 mm quadcopter came in completely electrically dead. The interesting part of a fault like that is not the fix — it is not jumping to a conclusion before the evidence supports one.

I worked through three sequential checks, each one eliminating a layer: the battery via multimeter, the PCB via USB-C, and then MOSFET continuity. That last check found burnt MOSFETs, and the root cause behind them was reverse polarity on a board with no reverse current protection designed in.

So the real failure was not the MOSFETs. It was a design that allowed a reversible connection to destroy the board — the components that burnt were doing exactly what unprotected components do.

A month of hands-on aircraft maintenance at Sarsan Aviation Academy in Dubai — the kind of exposure most engineering students do not get until years into a career.

I disassembled and reassembled an APU, a Soloviev D-30 turbofan, 737 and Piper landing gear, disc brakes, hydraulic tanks, fuel tanks, and a piston engine on decommissioned airframes. Taking a turbofan apart and putting it back together changes how you read a cutaway diagram permanently.

I progressed to leading a five-student team during the placement, working within a regulated environment where the safety and quality standards are not negotiable and the paperwork is part of the job rather than an afterthought.

For the Global Engineering Challenge, our six-person team designed ANVILS — Accurate Networked Vehicle Information Live Service — a UWB mesh network for real-time bus tracking across Sheffield's public transport network.

The system reaches 17 cm positional accuracy, roughly 17.5× better than what current tracking systems achieve. I modelled TDoA and Time of Arrival positioning across 54 receiver nodes along the 120 bus route.

The engineering only matters if the economics work, so I also ran the cost-benefit analysis: £670,000 implementation cost against a 2.3-year payback period, with 10 kWh/day energy consumption across 18 buses, targeting a 9.2% to 25% modal shift over five years.

I independently completed a Formula Student electrical recruitment brief in KiCad, laying out a PCB for a three-stage analog signal-conditioning circuit: amplifier, band-pass filter, and comparator.

I routed every trace manually rather than leaning on the autorouter. On an analog signal chain that is the right call — trace length and return path placement affect the signal in ways an autorouter optimising purely for connection does not account for.

The board passed both ERC and DRC checks with no functional errors.

I developed and programmed a solo embedded C++ reaction-time system using six LEDs and two buttons, iterating the code over two days.

The problem that actually mattered was CPU hammering. A busy-wait loop polling as fast as it can looks like it should give the most precise timing and does the opposite — it starves everything else and the measurements drift. Identifying that as the source of the timing inaccuracy was the real work; the fix followed easily once the cause was clear.

I validated the finished system against measured reaction times of 200–400 ms, which is the correct human range and confirmed the timing was sound.

I competed in two engineering competitions at Emirates Aviation University in Dubai.

In the team drone-building challenge I programmed the drone using a gyroscope and BetaFlight, tuning the flight stability. The team placed 1st.

In the water rocket challenge I was responsible for balancing the rocket's centre of mass against its centre of pressure to keep the flight stable — the same stability-margin principle I would later apply properly on TNO. The team placed 4th.