Engineering Internship at Parry Labs

I spent the summer embedded on the manufacturing floor at Parry Labs, designing fixtures, diagnosing failure modes, and tightening the flow of work when the team was racing to ship the first Defender units. It was a crash course in building for reality. I learned to design for technicians’ hands, to validate ideas quickly with prints that were ready by the next hour, and to communicate evidence clearly enough that mechanical engineering could act on it with confidence. The work pushed my design instincts toward speed with rigor, my problem solving toward structured curiosity, and my professional habits toward ownership, clarity, and momentum.

Fixtures to made precision fast

A gasket application fixture became my clearest example of design that changes the floor. Late in the program, the Defender’s water testing drove the addition of thin gaskets in tight spaces. Hand placement was slow and failure prone. Initial fixture concepts that supported the gasket perimeter fell apart the moment the liner was peeled; the liner was stiffer than the gasket and would pull the part off its supports, leaving a sticky mess.

Moving quickly through CAD and prints, I pivoted to a guided placement approach. The gasket had nine screw holes. I built a rigid frame with slender alignment pillars that extended from well outside the part envelope to each hole. The pillars registered the gasket precisely, held its geometry during liner removal, and drove a single smooth placement motion that seated every hole and edge in one pass. The fixture used only the two in-house printers and scrap hardware, which kept iteration loops tight and costs near zero.

The impact was immediate. Technicians could set the gasket, peel, and place in seconds without rework. Breakage dropped, supply requests fell, and throughput rose. The fact that it was the tool people asked for by name when building plenums told me the design hit the right balance of precision, speed, and simplicity. Behind that outcome were dozens of micro iterations. I printed variants with different pillar stiffness, surface finishes, and hole tolerances until the motion felt natural and repeatable. I learned to measure success not just by dimensional fit but by how confidently a tech could use the tool on the fifth unit of a long shift.

Using FEA to drive design

The operation control unit’s perimeter gasket was failing water tests because it was over compressed. The specification allowed up to 30 percent compression. We were seeing 40 to 45 percent in practice. Dimensional checks alone could not explain it. The pattern of indentation on the installed gasket hinted at asymmetric forces at four mounting points, which suggested warping and unintended loads from adjacent hardware.

I modeled the gasket in SolidWorks Simulation and built a parallel model of the plastic cover and battery door interface. By applying realistic closure forces and boundary conditions, I replicated the compression pattern and traced the root cause to a combination of cover warping and screw preload that exceeded assumptions. That analysis gave mechanical engineering a clear path forward. Increasing a critical width and controlling installation torque reduced unintended deformation and kept the gasket within its compression envelope. My role was to isolate the mechanism, show the force path, and quantify the margin. FEA was the bridge between what the floor was seeing and what the drawings needed to reflect, and it was the tool that let a manufacturing intern make a design conversation simple and actionable.

Implementing Lean manufacturing

When I arrived, fewer than one in five top level assemblies were complete and the team was one month from a major shipment. We did not have the luxury of perfect kits or clean handoffs. Lean for us meant ruthless clarity and forward progress on partial information. I set up a kitting and staging system that treated every manufacturing order as either partial or complete, with standardized locations and clear criteria for movement to the next step. I took ownership of the paper trail, assembling the instructions and lot tracking at the start of each job so technicians could spend time building rather than hunting documents.

We pushed work breakdown from assemblies to sub assemblies to components, attached realistic time estimates, and instrumented the floor with visible status on screens. The effect was to convert uncertainty into queues and time boxes. When parts arrived late, we could fold them into builds that were already two thirds done. When debate with engineering changed drawings, we had controlled locations to absorb the change. The team shipped the first 20 units on time. More importantly, the floor felt less like controlled chaos and more like a system that could absorb variation without losing momentum.

Problem solving grounded in fast feedback

The way I problem solve changed. I began with immediate hypotheses and then forced them into structured maps of possible causes, evidence, and tests. On the gasket compression issue, that meant drafting a cause chart from chassis geometry, hinge behavior, battery tolerances, screw preload, and cover deformation, then pruning it quickly with targeted checks and FEA. On fixtures, it meant testing motions on the floor first and designing the geometry to feel effortless under repeat use. I learned when to seek input. If I had deep familiarity with the task, I could iterate alone at high speed. If I lacked domain context, talking with senior technicians for ten minutes often saved a day of prints. The most valuable ideas often came as simple comments from people with thousands of hours of practice, and my job was to translate those into geometry and process.

Rapid prototyping was the backbone. I printed more than fifty fixture variants in the first week. That volume taught me to respect tolerances, to design for the way parts want to be handled, and to accept imperfect but valuable steps toward a working tool. Failing fast became information, not setback. I started to ask a different first question when picking up a new challenge. What is the fastest path to a result that the floor can use today, and how do I collect enough evidence along that path to make the next decision obvious.

Professional skills that made a difference

Two habits had outsized impact. Ownership was one. When something needed doing, I took it and finished it, whether it was modeling compression, building a fixture, or assembling paperwork. The other was communication. I learned to present findings with a crisp problem, clear evidence, and one recommendation. That made collaboration smoother and cut down on back and forth when time was tight. I also became more deliberate about user experience in hardware design. If a tool was even slightly awkward, adoption suffered. Designing for grip, motion, visibility, and error resistance turned out to be as important as the nominal dimensions.

The environment rewarded initiative. My managers gave me latitude to run with problems. Technicians gave me candor about what actually worked. Engineers appreciated analysis that matched real behavior. That mix taught me to bridge hands on work with analysis in a way that respects both.

What I will bring forward

This summer made me faster and more grounded as a designer. I can turn an idea into a tested tool in a day. I can use simulation to simplify complex mechanical behavior into clear decisions. I can build lean systems that keep teams moving when uncertainty is high. Most of all, I am comfortable owning ambiguous problems and making them concrete with evidence that others can act on. As I move deeper into aerospace, I plan to keep designing for the people doing the work, validating quickly, and communicating in ways that keep programs shipping without compromising quality.