Collin Coe

I'm a technology leader with over a decade of experience bridging the gap between complex technical challenges and real-world business needs. Currently at NetWorks Group, I focus on modernizing legacy infrastructure, implementing scalable systems, and leading technical initiatives that move the organization forward. My day-to-day involves architecting systems that need to survive real-world conditions — traffic spikes, security incidents, hardware failures, and the occasional human error. I spend a lot of time thinking about resilience: how do you build systems that do not just work, but keep working when things go wrong?

My background spans cybersecurity, cloud infrastructure, process improvement, and team leadership. I've managed enterprise-scale deployments, led cross-functional teams through major transformations, and built resilient systems that keep critical operations running. What excites me most right now is shaping how organizations integrate AI into their workflows — responsibly, effectively, and with clear purpose. AI is not a magic wand, and treating it as one leads to expensive failures. The organizations that succeed with AI are the ones that approach it with clear use cases, honest evaluation of what models can and cannot do, and a willingness to build the plumbing that connects models to actual business processes.

I believe in leading from the front: staying hands-on with technology while empowering teams to do their best work. Whether it's architecting infrastructure, optimizing processes, or mentoring the next generation of technologists, I'm driven by the belief that good systems and good leadership compound over time. The best technical decisions are ones that make the next decision easier — choosing the right abstraction, paying down tech debt before it compounds, documenting the things only one person knows. Leadership is the same way: investing in people pays returns long after the immediate project is done.

Outside of work, I'm building The Coe Lab — a personal platform for exploring AI, automation, and the intersection of technology and everyday life. The Coe Lab started as a place to write about what I was learning and has grown into a collection of self-hosted services, interactive experiments, and writing about technology. It is part portfolio, part playground, part public notebook. When I'm not at the keyboard, I'm with my wife and daughter, exploring the outdoors, or planning the next experiment.

Security-first. Every system I architect starts with the question: what happens if this gets attacked? That does not mean paranoia — it means defaults. Encryption at rest and in transit. Least privilege access. Audit logs that actually get reviewed. The goal is not to make systems impenetrable (nothing is) but to make breaches detectable, containable, and recoverable. A security-first mindset also means saying no to shortcuts that would save time now and create risk later.

Automation-focused. If I do something manually more than twice, it becomes a script. If a team does something manually more than once a week, it becomes a pipeline. Automation is not about eliminating jobs — it is about eliminating the repetitive, error-prone tasks that drain energy and attention from the work that actually requires human judgment. The best automation is invisible: people do not notice it is there, they just notice that things work.

AI-curious. I am genuinely excited about what large language models and AI tooling can do for organizations — and equally aware of their limitations. The hype cycle has created unrealistic expectations, but underneath the hype there are real capabilities that can transform how teams work. My approach is pragmatic: identify a concrete problem, test whether AI actually solves it better than the alternatives, and measure the results. If it works, ship it. If it does not, move on.

Built to last. Technology choices should account for the person who maintains the system two years from now — who might be you, might be someone else, and deserves clear documentation, sensible defaults, and an architecture that does not require tribal knowledge to operate. I prefer boring, proven solutions over exciting, unproven ones when the cost of failure is high.

My homelab is where work and hobby collide. A dedicated Unraid server runs Docker containers for AI tools, media management, remote desktop access, and monitoring. Everything is self-hosted, which means I am also the on-call engineer — when a certificate expires at midnight, I am the one who fixes it. This is not a burden; it is the point. Staying hands-on with infrastructure keeps my skills sharp and gives me a sandbox to test ideas before proposing them at work.

The most significant side project is OpenClaw, an autonomous AI assistant that runs in Docker and handles tasks ranging from home automation to traffic reporting. Building it has been a crash course in prompt engineering, API integration, and the realities of running AI systems in production — they are powerful but unpredictable, and managing that unpredictability is the actual engineering challenge.

I also build small web experiments — typing games for my daughter, reaction time tests, memory games. These live in the Playground section of this site and serve as a reminder that technology does not always need to be serious. Sometimes the best way to learn a new framework is to build something silly with it.