Mechanical engineering internship
Fulfil Solutions
Mechanical design and analysis for automated grocery fulfillment, with a focus on compliance, tolerance variation, and reliable material handling.
- Role
- Mechanical Engineering Intern
- Context / status
- May–August 2025
- Tools & methods
- SolidWorks · Spring analysis · Spreadsheet modeling · Physical prototyping · DFM
01
Overview
A note on project details: Much of my work is confidential, so I can’t share specific designs, prototype photos, internal calculations, or test results. This page focuses on the engineering methods I used and the skills I developed.
Fulfil builds robotic systems for online grocery fulfillment. Its public platform combines inventory storage, automated dispensing, mobile ShopBot robots, and temperature-controlled areas to pick and pack orders directly into grocery bags.
During summer 2025, I worked on mechanical hardware supporting bag handling and tote positioning. These tasks sit at the interface between flexible packaging and repeatable automation: the hardware has to accommodate variation while fitting into a larger moving system.
Company background: Fulfil’s public technology overview ↗
02
My contribution
I compared spring-based mechanisms, built a spreadsheet to evaluate deflection and force, and spent a large part of the summer making and testing physical prototypes. I sourced spring options through OSH Cut and McMaster-Carr, revised CAD for injection molding and CNC machining, and passed the designs and supporting work back to the team. I also led mechanical concept work for a positioning mechanism and coordinated electrical requirements and test planning.
03
Design & development
Spring selection and compliance
I considered torsion springs, beam-bending springs, and linear springs to provide controlled motion and contact force. Each offered a different balance of travel, stiffness, packaging, and assembly. The goal was a useful operating deflection across dimensional variation, with enough force to hold the bag securely without damaging it.
Spring equations and design tradeoffs
The equations below illustrate standard idealized models. They do not reproduce internal designs, spreadsheet values, or test data.
LINEAR SPRINGS
F = F0 + kδ
Force F increases with displacement δ at spring rate k; F0 represents any initial preload. This made the connection between tolerance-driven travel and changing contact force explicit. A lower rate reduces force variation over a given travel, while preload changes the starting force. This model applies over an approximately linear working range.
TORSION SPRINGS
T = T0 + kθθ · F = T / r⊥
A torsion spring provides torque T as its angle θ changes. T0 is preload torque, kθ is angular stiffness, and r⊥ is the perpendicular moment arm of the contact force. I considered how rotation and leverage translated spring torque into useful force; that leverage can change as a mechanism moves. Angular units must match the spring-rate units.
BEAM-BENDING SPRINGS
δ = FL3 / (3EI) · k = 3EI / L3
I = bt3 / 12 · σmax = 6FL / (bt2)
For a uniform rectangular cantilever with a transverse tip load, L is free length, E is Young’s modulus, I is the second moment of area, b is width, and t is thickness in the bending direction. These small-deflection, linear-elastic relations show why thickness and free length strongly affect stiffness, and why deflection needs to be checked alongside peak bending stress. Different supports and loading require different beam models.
General references: Lee Spring: torsion springs ↗ · NASA: mechanical design reference ↗
Spreadsheet analysis and working deflection
The spreadsheet was a central part of the design process. I used it to connect assembly tolerance stackups with required travel, then compare spring options by their predicted force, deflection, and stress. It let me change assumptions and see how a candidate behaved across the working range, rather than selecting a spring from one nominal position.
I used those comparisons to identify a practical deflection range: enough compliance to accommodate variation, with suitable contact force and room for the mechanism to move. “Optimal” meant balancing those requirements. The spreadsheet narrowed the options; physical prototypes were how I checked whether the predicted behavior held up in the assembly.
Spring sourcing and physical prototyping
I sourced springs through OSH Cut and McMaster-Carr and did extensive hands-on building, fitting, and testing. Trying the hardware helped confirm useful deflection and exposed contact, clearance, and assembly issues that the calculations could not fully capture.
I fed those observations back into spring selection, geometry, and the spreadsheet. That repeated cycle of calculation, build, test, and revision was a substantial part of my contribution, and helped establish which options were worth developing further.
Design for manufacturing and team handoff
I also worked through DFM on multiple designs for injection molding and CNC machining. That meant revising the CAD with the intended process in mind and considering how parts would be made, assembled, and integrated—not just whether the mechanism moved correctly.
I passed the revised designs and supporting analysis back to the team, documenting the reasoning behind changes, prototype findings, manufacturing considerations, and remaining questions. The handoff gave the team a basis for continuing development after my internship.
Positioning and actuation
For the positioning work, I compared linkage and actuator approaches while considering load paths, alignment, packaging, and behavior when power was removed. I coordinated those mechanical choices with the electrical and testing work.
04
Results & next steps
I handed off spring comparisons, the analysis spreadsheet, CAD revised for manufacturing, and findings from physical prototypes. The work connected calculations to hardware decisions and gave the team a documented starting point for continued development.
Further system testing, final design selection, and production-readiness work remained open at the end of the internship.