Description

A student sits down to learn and the app won’t load. A guide’s dashboard is showing the wrong mastery scores. L1 and the AI have already tried and failed. Now it is yours — and a kid’s learning day is waiting on you to figure out what is actually wrong.

Most of your day goes to the tickets nobody else could close, across the many apps that make up Alpha’s learning platform. You reproduce the failure in an environment that mirrors the student’s or guide’s setup, dig through logs and docs, put AI to work to move faster, and then verify everything it gives you, because a wrong answer sent into a classroom compounds quickly. You will ramp up on a new product almost every week. That is the job, not a warning.

You own each problem end-to-end: reproduce it, find the root cause, and resolve it in as few touches as it takes to get the student or guide moving again. The reasoning you write down makes the next agent and the next AI workflow smarter, so your fix keeps paying off long after you move to the next ticket. Over time, you become the person the team sends its hardest education-product problems to.

If you want the hard, ambiguous problem, and you actually care that the product on the other end helps a kid learn, apply.

What you will be doing

Resolve ambiguous, escalated tickets from students, parents, and guides that AI and L1 support could not close.
Reproduce failures in an environment that mirrors the user’s device, app, and configuration, using logs and real artifacts.
Investigate across Alpha’s learning apps (tickets, Slack, knowledge bases, logs) before escalating.
Direct AI tools (such as ChatGPT or Claude) to speed up diagnosis, ground them in real documentation, and verify every output.
Communicate clearly and calmly with non-technical users, gathering enough up front to resolve in a single pass.
Route issues to engineering with full diagnostic context when a product is genuinely broken.
Document your reasoning so the next person, and the next AI workflow, can reuse it.

What you will NOT be doing

Following scripts or decision trees on a single product.
Handling the easy queue. AI already closes the tickets a knowledge-base lookup can solve.
Forwarding hard problems to engineering without first reproducing and narrowing them.
Letting AI think for you, or pasting its output without checking it.
Waiting to be told what to do, or for someone else to unblock you.

Key responsibilities

Resolve complex, ambiguous customer issues to root cause across Alpha’s education products.

Candidate requirements

2+ years in a hands-on technical role such as technical support, software engineering, QA, sysadmin, or DevOps. The title does not need to be “support.”
Comfortable making and reading REST API calls and JSON, interpreting HTTP status codes (such as 401 vs 404 and 429 vs 403), and working in a command line and logs.
Hands-on experience using AI tools (such as ChatGPT or Claude) in your daily technical work.
Professional fluency in English, written and spoken.
Available to work full-time (40 hours/week), with all working hours falling between 1:00 PM – 10:00 PM UTC (8:00 AM – 5:00 PM US Eastern).

Nice to have

Experience supporting non-technical end users (in education, edtech, or consumer apps), not just IT professionals.
Hands-on experience directing AI tools and catching their mistakes.
A track record of troubleshooting across several unrelated products, not just one.
Developer-grade depth: you can read code or trace an API call when the problem demands it.

To apply for this job please visit ed.crossover.com.