About Kempbell Consulting

I'm Travis Kempbell, a Washington-based principal engineer with 15 years of experience. I help SaaS teams build AI features that work in production, and I keep the infrastructure underneath them stable and affordable. Kempbell Consulting is my practice. All work is delivered by me.

Travis Kempbell giving two thumbs up on a snowy mountain ridge

Background

After fifteen years with growth-stage SaaS companies, I've seen the same pattern repeatedly: as teams grow from early product-market fit to serious revenue, infrastructure lags behind. Deployments get slower, AWS bills grow faster than usage, and engineers end up doing ops work instead of building. More recently the same pattern shows up in AI features: demos that work in a meeting, then fall over under production traffic, latency requirements, or their own inference costs.

I've fixed both. On the infrastructure side: cutting AWS costs by six figures, rebuilding CI/CD pipelines to take deploys from hours to minutes, and designing systems that scale without constant firefighting. On the AI side, I build and operate my own production ML systems, including a custom speech-recognition engine and a live LLM-based SaaS. I run these systems myself, so the advice is based on direct experience.

The case studies show this work in detail.

How I Work

I build infrastructure that gets out of your way. I choose technologies with years of production history over new and unproven ones; where possible, infrastructure should be boring. Manual processes create risk, so I build systems that deploy, scale, and recover without a person in the loop.

I benchmark before and after. Model migrations and cost changes get verified with real evaluation before rollout, and everything I leave behind is documented well enough that your team can understand and maintain it without me.

Beyond Infrastructure

Infrastructure is the foundation, but I also build what runs on it:

  • AI features built for production: LLM integration, structured extraction, and semantic search, with latency budgets, cost controls, and evaluation before release.
  • Internal dashboards and tools in Next.js, React, and TypeScript that give teams visibility into their systems.
  • Backend services that connect infrastructure to application code.

Why Work With Me

I work as a senior technical lead embedded with your team, and the engagement is done when the problem is fixed. Everything I ship runs in production and comes with a written record of what changed.

My practice does not depend on any one client, so you get an honest read on your situation, including when the honest read is that you do not need me.

When work calls for skills beyond mine, I bring in trusted specialists and remain accountable for the result.

Need faster deployments, lower AWS bills, or an AI feature that works in production? Let's fix it.