When I started Merch Masters LK, my 3D printing gift business, during the COVID pandemic, I never expected it would profoundly influence my professional career as a DevOps engineer. Yet as I spent countless hours calibrating printers, tweaking designs, and optimizing workflows, I began to recognize striking parallels between these two seemingly different disciplines. Both demand a precise blend of technical knowledge, process optimization, and continuous improvement—skills that have become invaluable in my work at the London Stock Exchange Group.

Iterative Development and the Perfect Print

In DevOps, we embrace iterative development over the waterfall approach. Small, incremental changes allow for faster feedback and adjustment. 3D printing taught me this principle in a tangible way that digital systems simply can't.

When I started designing custom gifts for Merch Masters LK, I quickly learned that perfect prints rarely happen on the first try. Even with a meticulously crafted 3D model, factors like material properties, printer calibration, and environmental conditions all influence the outcome. Getting it right requires iteration.

For example, when developing a custom phone holder for a client, I went through five iterations:

  • Iteration 1: Basic design - functional but with support structure issues
  • Iteration 2: Adjusted support angles - better but exhibited stringing
  • Iteration 3: Adjusted retraction settings - improved quality but with dimensional inaccuracies
  • Iteration 4: Calibrated flow rate - better dimensions but insufficient strength
  • Iteration 5: Changed infill pattern and density - finally achieved the perfect balance

This experience directly mapped to my approach to CI/CD pipeline development at LSEG. Rather than designing an entire pipeline as a monolithic entity, I started implementing small, testable components that could be validated incrementally. This reduced troubleshooting time and prevented cascading failures.

Configuration as Code and Print Profiles

Infrastructure as Code (IaC) is fundamental to modern DevOps. By defining infrastructure through code, we ensure consistency, enable version control, and facilitate automation. My experience with slicing software and print profiles provided a perfect analog to this principle.

Each 3D printer configuration contains hundreds of parameters: 

; Example portion of a PrusaSlicer configuration file
[printer:Ender3Pro_PLA]
bed_temperature = 60
bridge_fan_speed = 100
compatible_printers_condition = printer_notes=~/.*PRINTER_VENDOR_CREALITY.*/
default_filament_profile = Generic PLA
extruder_temperature = 200
fan_always_on = 1
filament_cooling_moves = 4
first_layer_bed_temperature = 65
first_layer_temperature = 210
...

Managing these configurations manually would be chaotic. Instead, I created a library of baseline configurations (much like Terraform modules) that I could extend and customize for specific print requirements. These configurations were stored in version control, allowing me to track changes, roll back when needed, and share optimal settings across multiple printers.

This discipline translated directly to my work on the DX1 Uploader automation project at LSEG, where I applied the same principles to infrastructure definitions. I developed a custom Terraform library for AWS MSK clusters that encapsulated best practices while allowing for environment-specific customization. Just as with print profiles, these modules were versioned, tested, and shared across teams.

Failure Recovery and Resilience

In both 3D printing and cloud infrastructure, failures are inevitable. The difference between success and disaster is how quickly you can detect, diagnose, and recover from these failures.

3D printing taught me hard lessons about failure. A 12-hour print that fails in hour 11 is painful—especially when a customer is waiting. This forced me to develop robust monitoring and recovery strategies:

  • Using cameras to monitor prints remotely
  • Building recovery procedures for common failure modes
  • Implementing preventative maintenance schedules
  • Creating detailed checklists for printer setup before long jobs

I found myself applying these exact same principles when directing the RHEL 8.10 AMI upgrade across our production environments at LSEG. Just as I wouldn't start a 12-hour print without a thorough pre-flight check, I wouldn't deploy to production without:

  • Comprehensive pre-deployment testing in staging environments
  • Detailed rollback procedures for each component
  • Monitoring alerts configured to detect anomalies early
  • Clear communication channels for all stakeholders

Resource Optimization: Filament and AWS Usage

Both 3D printing and cloud infrastructure involve managing finite resources, and efficiency directly impacts the bottom line.

When running Merch Masters LK as a business, I became acutely aware of how every gram of filament, kilowatt of electricity, and minute of print time affected my margins. This drove me to optimize aggressively:

  • Nesting multiple parts to maximize bed usage
  • Adjusting infill density based on structural requirements (not everything needs 100% infill)
  • Fine-tuning print speeds to balance quality and production time
  • Scheduling large prints during off-peak electricity hours

These optimization habits proved invaluable when managing cloud resources at LSEG. I found myself naturally looking for inefficiencies in our AWS deployments:

  • Implementing auto-scaling to match resource allocation with actual demand
  • Using spot instances for batch processing jobs
  • Setting up lifecycle policies to archive infrequently accessed data
  • Scheduling non-critical tasks during lower-cost time periods

In both contexts, the principle is the same: measure carefully, optimize deliberately, and always question if there's a more efficient approach.

Automated Testing and Print Calibration

Automated testing is a cornerstone of DevOps, enabling quick validation of changes before they impact production. The 3D printing equivalent exists in calibration prints and testing procedures.

When setting up a new printer or filament, I don't immediately jump to complex production prints. Instead, I run a series of standardized calibration prints:

  • Temperature towers to identify optimal extrusion temperature
  • Retraction tests to minimize stringing
  • Bridging tests to verify cooling settings
  • Dimensional accuracy tests to ensure precise sizing

These calibration prints function exactly like unit and integration tests in software development. They isolate specific aspects of printer performance, allow for targeted adjustments, and provide a baseline for measuring improvements.

I applied this testing mindset when enhancing Jenkins pipelines at LSEG. Rather than making changes directly to production pipelines, I created a test framework that validated pipeline components using: 

// Example Jenkins pipeline testing snippet
stage('Test Pipeline Component') {
    steps {
        script {
            // First verify the behavior of the component in isolation
            runComponentTest('deploymentModule')
            
            // Then verify integration with dependent systems
            runIntegrationTest('deploymentModule', 'monitoringSystem')
            
            // Finally verify with simulated production data
            runSimulationTest('deploymentModule', 'productionDataSet1')
        }
    }
}

This systematic approach to testing, inspired by my 3D printing calibration routine, dramatically reduced production issues and accelerated our pipeline development cycle.

Documentation and Knowledge Sharing

Perhaps the most valuable lesson from running Merch Masters LK was the critical importance of documentation and knowledge sharing.

When operating multiple printers with different materials, I quickly discovered that relying on memory alone was insufficient. I needed a system to track what worked, what didn't, and why. I created detailed logs for each printer:

  • Maintenance history and component replacements
  • Optimal settings for different material types
  • Common failure modes and resolution steps
  • Design modifications for improved printability

This documentation became invaluable as my business grew, enabling consistent quality and faster onboarding when I needed help during peak periods.

I've since applied this same documentation rigor to my DevOps work. When I developed the custom Terraform library for AWS MSK clusters at LSEG, I made sure to create:

  • Detailed README files with usage examples
  • Architecture diagrams explaining component relationships
  • Troubleshooting guides for common errors
  • Decision logs explaining why certain approaches were chosen

This documentation transformed the library from a tool that only I could use effectively into a resource that empowered the entire team, significantly reducing our collective time to deployment.

Conclusion: The Maker Mindset in DevOps

Running Merch Masters LK alongside my DevOps career created a powerful feedback loop of skills and insights. The hands-on, physical nature of 3D printing made abstract DevOps concepts tangible, while my software engineering background helped me approach 3D printing with a systems mindset.

The most valuable lesson was developing what I call the "maker mindset": a blend of curiosity, pragmatism, and persistence that drives continuous improvement in both fields:

  • Curiosity to explore new technologies and techniques
  • Pragmatism to focus on results rather than perfection
  • Persistence to overcome inevitable failures

This mindset has proven invaluable throughout my DevOps journey at LSEG, particularly when leading complex projects like the DX1 Uploader automation and RHEL 8.10 AMI upgrade.

If you're a DevOps professional looking to expand your skills and perspective, I highly recommend exploring hands-on maker hobbies like 3D printing. The tangible feedback loop and physical manifestation of your decisions create learning opportunities that purely digital work cannot provide.

And if you're a 3D printing enthusiast considering a career in tech, know that your maker skills translate remarkably well to DevOps. The problem-solving abilities, system optimization experience, and resilience you've developed at the printer bed are exactly what modern IT organizations need.

What maker skills have you found valuable in your technical career? I'd love to hear your experiences in the comments below.