DevOps / Integrations Engineer — AI Client Instances

Please read the schedule requirement before applying: this role requires working 9:00 AM to 5:00 PM Pacific Time, Monday through Friday, plus every other Saturday on the same 9 to 5 Pacific hours. You'll be supporting our team and our clients in real time, so simply overlapping a few hours is not enough — you must be fully available and online on this exact schedule, consistently. This is absolutely non-negotiable. If you cannot commit to Monday through Friday plus alternating Saturdays, 9 to 5 Pacific, this is not the role for you. You will also need a multi-monitor setup with two or more screens, and you'll be asked to show proof of it during the hiring process. We do not work with people running off a single laptop screen — the work moves too fast and involves too much at once for that. About the work: we run an AI operating system that we deploy as isolated instances for each of our clients. We're hiring a dedicated developer to own the full instance lifecycle — spinning up new client environments, wiring in their integrations and communication channels, building and configuring agents, troubleshooting live issues, and documenting everything so it becomes repeatable. This is a hands-on, client-facing, team-embedded role. You'll work hand in hand with our staff and directly with our customers — developing, integrating, and supporting side by side — which is exactly why strong, clear English communication matters just as much as your technical skills. The way we build is non-negotiable and I want to be upfront about it: everything we do is done through AI-assisted development using Claude Code and/or OpenAI Codex. There is no other way we work, so you must have real hands-on experience with at least one of these tools. You don't need to be an expert in both, but you do need to already work this way. Equally important, we never hard-code tokens, keys, or credentials — everything runs through proper, dynamic secrets management. If you're an old-school developer who's used to dropping API keys into code, this isn't the right fit. I've spent the last year learning that developers with those old habits slow everything down, and I need someone who already works the modern, clean way and can hit the ground running. Two more requirements that matter to us. First, this is an exclusive engagement — while you're working with us, you are not to be working with any other companies or clients at the same time. We need your full focus. Second, we live on Slack and we expect fast responses — not 15 or 20 minutes later. If you're at your computer working, as you should be during your hours, you should be able to reply within a couple of minutes. Quick, reliable communication is a core part of this job, not a nice-to-have. Day to day, you'll create new client instances by cloning or creating the repo from our template, deploying it on a platform like Render, connecting GitHub, and configuring environment variables, API keys, and tokens the right way. You'll manage DNS, subdomains, and tunnels in Cloudflare and point public names to the correct services. You'll configure AI providers such as OpenAI and Anthropic Claude for each client, including both client-owned and managed keys. You'll build and configure agents, MCP servers, and extensions, and keep track of what's active for each client. You'll integrate communication channels such as Slack, Twilio, Telegram, WhatsApp, Signal, email over IMAP and SMTP, Discord, and webhooks based on each client's needs. You'll verify every deployment end to end before handoff, troubleshoot issues live with clients, and document as you go so that every fix becomes a written procedure and every repeated task becomes a reusable template. You'll also keep instances healthy with backups, logging, and recovery. To be a strong fit you must have: availability 9 to 5 Pacific Monday through Friday plus every other Saturday; a two-or-more monitor setup you can prove; excellent English communication for working directly with clients and our staff; hands-on experience with Claude Code and/or OpenAI Codex; strong secrets discipline with no hard-coded credentials; the ability to work with us exclusively while engaged; fast Slack responsiveness within minutes; hands-on experience deploying apps on a platform like Render, Railway, Fly.io, or Heroku; solid Cloudflare experience with DNS, subdomains, and tunnels; comfort with Git and GitHub and connecting repos to deployment pipelines; experience integrating third-party APIs and messaging channels; and familiarity with the OpenAI and/or Anthropic Claude APIs. Nice-to-haves include experience with MCP servers or agent frameworks, multi-tenant or per-client deployments, Docker and CI/CD, and auth platforms such as Clerk. When you apply, please answer these briefly: (1) Can you commit to working 9:00 AM to 5:00 PM Pacific Time, Monday through Friday plus every other Saturday, consistently? Please confirm your time zone. (2) Do you have a multi-monitor setup with two or more screens? (3) Are you able to work with us exclusively, with no other clients or companies during the engagement? (4) Which AI coding tools have you used — Claude Code, OpenAI Codex, or both — and how? (5) Describe a time you deployed and managed an app on Render or a similar platform. (6) Which messaging or API integrations have you built, and what was the trickiest part of one? (7) Tell me about a time you worked directly with a client to troubleshoot a technical problem.