Lance Keay

Senior full-stack developer and DevOps engineer. I like systems that are boring to operate. I build the application, the pipeline that ships it, and the monitoring that says when it's unwell.

Profile

I work across the whole delivery path. I write TypeScript and Python services and React front ends, then do the containers, infrastructure-as-code and CI/CD that get them to production and keep them there. Clients usually bring me in when a product has outgrown its first deployment: builds are slow, environments drift, nobody trusts the release process, or the cloud bill has crept up.

Lately a growing share of my work is putting AI tooling into real systems: model gateways, local and hosted LLM serving, agent workflows with proper permissions, and the cost tracking that stops a prototype becoming a surprise invoice. The reviews on this site come out of that work. I test the tools before I recommend them to anyone who's paying me.

  • Plain over clever

    Code and infrastructure the next person can read, change and roll back.

  • Measured, not guessed

    Metrics, logs and cost data before opinions, including mine.

  • Written down

    Runbooks, decision records and handover notes come as part of the job.

Skills

Languages

  • TypeScript
  • JavaScript
  • Python
  • Go
  • C#
  • SQL
  • Bash

Application

  • Node.js
  • React
  • Next.js
  • .NET
  • FastAPI
  • REST & GraphQL

Data

  • PostgreSQL
  • Redis
  • SQL Server
  • Message queues

DevOps & platform

  • Docker
  • Kubernetes
  • Helm
  • Terraform
  • Ansible
  • GitHub Actions
  • GitLab CI

Cloud & hosting

  • AWS
  • Azure
  • Proxmox
  • Linux
  • Nginx / Caddy

Observability & AI

  • Prometheus
  • Grafana
  • Loki
  • LLM serving
  • Agent tooling
  • Evals & cost tracking

Experience summary Sample

Placeholder summary to show the layout. Lance's real history is in the downloadable CV.

  1. Recent years

    Contract full-stack & DevOps engineer

    Self-employed · remote and on-site

    Platform and delivery work for product teams: moving services onto Kubernetes, rebuilding CI/CD so a release is one reviewed merge, cutting cloud spend by right-sizing and removing idle resources, and adding LLM features behind a gateway with logging and budgets.

  2. Earlier

    Senior developer, then lead

    Permanent roles · product software

    Owned services end to end, from API design to on-call. Introduced infrastructure-as-code and automated test environments, and mentored developers on testing and deployment practice.

  3. Start

    Developer

    Web and line-of-business applications

    Full-stack web development on SQL-backed systems, where I learned that most incidents are deployment incidents.

Have a system that needs a steady pair of hands?

Send a short note with what you're running and what's hurting. I'll tell you plainly whether I'm the right fit.