Selected Work

Dylan
Pires.

Versatile mechanical engineer, recent MSc graduate from ETH Zürich, with broad interests and expertise in additive manufacturing, computer aided engineering, and product development.

A selection of relevant projects from the last couple of years, spanning experimental non-planar FDM printing, proof-of-concept work in the medical field, and metal AM for motorsport. Prepared as an application for the LightSpray Production Intern role at On.

Author Dylan Cernadela Pires MSc Mechanical Engineering, ETH Zürich

Contact [email protected] Zürich, Switzerland

For On - LightSpray Intern Application portfolio

Dylan Pires · Portfolio01

Cover Letter

Why On, Why Now

Why On, Why Now.

I have been curious about how things work, how they are made, and how they could be made better for as long as I can remember. That curiosity is what led me to mechanical engineering, and it is what kept me engaged throughout my studies at ETH, where I specialised in product development, computer aided engineering, and additive manufacturing.

My internship at EFalke, a startup developing electric boats, put that foundation into practice. In a small team working on new technology, every decision had to account for manufacturability and scalability from the start. What I took away is that good engineering does not stop at solving the technical problem. It means making the solution repeatable and ready for the people who come after you.

That is exactly why the possibility of working on the LightSpray technology excites me. Helping refine a production method that cuts CO₂ emissions by 75% and creates an upper in just three minutes is the kind of work I want to be doing. I want to be the person on the team who gets under the hood, spots what is slowing things down, and finds practical improvements, whether that means supporting the data cockpit, helping refine SOPs, or contributing to the rollout of new production workflows.

The timing feels right too. Having already experienced what it is like to work in a small startup, I now want to see how things work at a larger, international company, and On's internship programme seems like exactly the right bridge. The way it was described at the CareerFairy livestream, and the people I have seen representing On, made it clear this is a proper programme for growth on both sides: there is a real onboarding, meaningful projects with real responsibility from early on, and a whole cohort of interns starting at the same time. Being able to share that experience with people joining from different parts of the company sounds like a great way to feel part of the team from day one. I would love the chance to contribute to the team and I am confident I would be a great fit.

Dylan Pires · Portfolio02

Contents

Six projects

Selected work.

01

G-code Modulator Self-built tool for parametric toolpath modulation in FDM

Web · Python · CAM · FDM

02

EVLP Lung Chamber Proof-of-concept rotating chamber for ex-vivo lung perfusion

Medical · FDM · Product Dev

03

Kitchen Scale Redesign Industrial design exercise, KTH exchange semester

Industrial Design · FDM · Blender

04

Motor Cooling Case LPBF wheel carrier + cooling jacket, AMZ Racing

LPBF · NX · FEM · DfAM

05

Pop-Pop Boat Thermal-driven micro engine, LPBF

LPBF · NX · Thermo

06

Latrine Emptying Device Hands-on team project, low-resource sanitation

Product Dev · Granta · Group

Contents03

01

G-code Modulator.

A browser-based tool I built to manipulate sliced G-code parametrically, going one level deeper than what any slicer or CAM tool exposes today.

Slicers expose layer-level controls: wall counts, infill density, layer heights. They do not expose the toolpath itself. The G-code Modulator I built works one level deeper: it ingests sliced G-code and applies parametric transforms directly to the X/Y/Z coordinates of every move. Layer lines, which slicers and printer manufacturers usually treat as defects to be hidden, become a deliberate design feature you can shape. The tool is live at gcode.dylanpires.com.

Fig. 1.1 - The G-code Modulator. Sliders drive parametric transforms; the viewer rebuilds the toolpath in real time.

Fig. 1.2 - Example output #1

Fig. 1.3 - Example output #2

Fig. 1.4 - Example output #3

Fig. 1.5 - Functional part made with the tool: lampshade

Fig. 1.6 - Functional part made with the tool: plant pot

01 · G-code Modulator04

02

EVLP Lung Chamber.

A rotating chamber for ex-vivo lung perfusion, taken from first sketch to a working proof of concept through four FDM-printed design iterations in one semester.

Semester thesis at ETH PDZ. The brief: design a chamber that holds and rotates a donor lung during ex-vivo perfusion, to test whether changing orientation slows degradation. No existing solution to reference, so the design had to be developed from scratch.

Four FDM-printed iterations, each addressing the failure mode of the previous one. Each version was printed in-house, assembled, and tested before the next geometry was started. The fast print-test-redesign loop was what made it feasible as a semester project. Even at this early proof-of-concept stage, the design was developed with an eye on how it could eventually be manufactured at scale. The fourth iteration held and rotated a real porcine lung during simultaneous ventilation and perfusion, validating the concept.

Fig. 2.1 - V1 → V4. Each prototype was FDM-printed and tested; each failure mode informed the next geometry.

Fig. 2.2 - Printed chamber on test stand

Fig. 2.3 - V4 CAD with anatomical lung

Fig. 2.4 - Active ex-vivo test, porcine lung

02 · EVLP Chamber05

03

Kitchen Scale Redesign.

An industrial design exercise reimagining a kitchen scale through the visual and functional language of an existing brand. We chose Teenage Engineering.

Group project for an industrial design class during my exchange semester at KTH Stockholm. The brief: redesign a kitchen scale inside the visual and functional language of an existing brand. We chose Teenage Engineering, a Swedish design house defined by industrial honesty and precise mechanical detailing, and called the result KS-01 Heavy. After designing in Fusion 360 and visualising in Blender, we built the physical mockup using FDM 3D printing: housing printed at 1:1 scale, then sanded, primed, and finished to read as a real consumer product. The mockup was the most important deliverable, using FDM as the means to produce a tangible, presentable object rather than just a prototyping shortcut.

Fig. 3.1 - KS-01 Heavy in three colour variants, each showing a different menu state. Rendered in Blender.

Fig. 3.2 - Knurled aluminium scroll wheel

Fig. 3.3 - Recessed on/off slider

Fig. 3.4 - The 1:1 FDM-printed mockup of KS-01 Heavy.

03 · Kitchen Scale06

04

Motor Cooling Case.

Consolidating the wheel carrier and motor cooling jacket into a single LPBF part for AMZ Racing, a piece that only exists because metal AM allows it.

A group project for the ETH "Design for AM" lecture in collaboration with AMZ Racing. The goal: combine two traditionally separate parts, the wheel carrier and the motor cooling jacket, into a single load-bearing LPBF component, with coolant channels routed through the structural geometry. The final design uses an internal lattice structure (not visible in the renders) to lightweight the part. My contribution was the FEM stiffness validation in Siemens NX: confirming that the consolidated, lightweighted carrier still met load-path requirements under cornering load cases.

Fig. 4.1 - FEM displacement under cornering load.

Fig. 4.2 - Printed LPBF component, post-process

04 · Motor Cooling Case07

05

Pop-Pop Boat.

A thermodynamic toy as a vehicle for learning the practical constraints of LPBF.

A pop-pop boat is a thermoacoustic toy: it boils a small water charge inside a coil, the steam pushes water out the rear, condensation pulls it back in, and the cycle repeats at the system's natural frequency to produce oscillating thrust.

Group project for the ETH "Design for AM" lecture. The goal was to learn the principles of design for AM by designing the internal coil geometry within LPBF rules: minimum channel diameter, unsupported overhang limits, and build orientation all shaped the final form.

Fig. 5.1 - External CAD. Cone profile chosen for LPBF orientation.

Fig. 5.2 - Internal section. The coil is the entire reason the part exists.

Fig. 5.3 - Printed component. LPBF surface texture without post-processing.

05 · Pop-Pop Boat08

06

Latrine Emptying Device.

A hand-powered, low-cost device for emptying pit latrines in resource-poor settings, a team project where I got my hands dirty across design, fabrication, and field testing.

Group project for the ETH "Product Development & Engineering Design" lecture, with the brief to create a device for emptying pit latrines safely, cheaply, and without grid power. Beyond the device itself, the project was a theoretical exercise on the full product lifecycle: development, manufacturing with locally sourced low cost materials, distribution, usage in the field, and disposal, considering every step from extraction through to end of life. As a small team we were all involved across the whole project, from the mechanism design and material selection in Ansys Granta to the hands-on building and testing.

Fig. 6.1 - Working sketch of the transport mechanism

Fig. 6.2 - Hands-on field testing

Fig. 6.3 - Team at final demonstration

06 · Latrine Device09

End

Thank you for reading

Let's build
something.

If anything here resonates with what you're doing at On, I would be very glad to talk further.

Dylan Cernadela Pires [email protected]

Web portfolio lightspray.dylanpires.com

Location Zürich, CH

Dylan Pires · Portfolio10

DylanPires.