Full Stack Applied AI Engineer - Backend & Cloud (work from home)

About HoloTutor

HoloTutor is an AI tutor and lesson platform for schools, colleges and training organizations, built by VictoryXR. Teachers create and assign lessons; students learn with an AI tutor that narrates, asks questions, and remembers how each learner is doing.

The Role

We’re looking for a Full Stack Applied AI Engineer who works across the whole product: the AI features, the Python API, the React app, the cloud it runs on, and the systems it connects to. You’ll join a small team that works closely together, brainstorming and building new features and systems every week, and you’ll help move ideas from first prototype to production. Your main focus is the back end. You’ll start by learning the platform alongside the team, then take over as the lead on our API, cloud infrastructure, database and integrations.

You apply AI in two ways. You build AI features that real teachers and students use, at a cost we can measure. And you build using AI: agents write much of our production code, and directing them, checking their work, and owning the results is what an engineer on HoloTutor is expected to do. You need to know what modern models can and cannot do, and how to make them reliable in a product.

Just as important, you’re comfortable on the back end. You know your way around cloud infrastructure and production databases, and you know how to change them without breaking them.

What You Will Do

• Build AI features. The tutor, text to speech, memory, and content systems. Decide which model calls what, and when.
• Keep AI cost and reliability under control. We meter every model, voice and image call.
• Ship across the full stack. APIs, database schemas and migrations, React interfaces, admin tools, dashboards and reports.
• Get the business logic right. Licenses, seats, subscriptions and orders. This code is money and customer trust, so it has to be correct.
• Connect systems. Integrate with VictoryXR’s other platforms, school LMSs (sign-in, rosters, assignments) and e-commerce. Write the API contracts other teams build against, and keep connected systems in sync.
• Run what you ship. Deploy it, migrate it, monitor it, and lead the fix when something breaks.
• Protect student data. Build with student privacy laws (FERPA, COPPA) in mind.

Infrastructure and Data

This is a big part of the job. We run on Google Cloud, and you’ll own a lot of it.

• Cloud hosting. Containers, managed databases, queues, storage, scheduled jobs, CI/CD. You know how these fit together and what they cost.
• Access and secrets. Least privilege, service accounts, secret management. You know why the guard rails exist and you don’t go around them.
• Databases. Schema design, indexes, query performance, and knowing when a query will hurt at scale.
• Migrations. Additive and safe to re-run, applied before the code that needs them, tested against real data before they touch production.
• Production safety. Backups and point-in-time recovery checked before risky changes. Staging first. Read before you write. A one-off script against prod gets the same care as a feature.
• Agents near production. AI tools are fast and confident, and that is the risk. You know how to give them access safely, and when not to.
• Observability. Logs, metrics and alerts that tell us something is wrong before a customer does.

How We Work

Agents write much of the code; your judgement is what makes it good. We brainstorm together every week, plan bigger work in writing before building, audit risky areas before they ship, and run several features in parallel. Everything goes to staging first, and production is a deliberate step. “Done” means verified in the real thing, not just a green build.

Technical Skills

Agents write much of our code, so what matters most is understanding: knowing these areas well enough to direct the work, review it, and catch what’s wrong.

• Strong understanding of full-stack development: Python back ends, TypeScript and React.
• Strong understanding of relational databases (MySQL or Postgres), with real production experience, including migrations on live data.
• Hands-on with a major cloud platform (GCP strongly preferred; AWS or Azure is fine): deploys, networking basics, IAM, logs and monitoring.
• Solid understanding of integrations: APIs, authentication and SSO, webhooks, and keeping connected systems in sync.
• Practical experience calling LLMs and oth…