Launch in
Role details for Founding Engineer
The person
- You are the person who does not trust a demo just because it looks smart.
- When everyone else is hyped by a demo, you are already asking: where does it fail, how often, can the user see the failure, and what should be checked by code instead of model judgment?
- You are not here to make AI sound magical. You are here to make it useful, inspectable, and safe enough for real teaching work.
What you will own
- You will own trust that is fundamental to the future of Classiqo.
- Classiqo will generate high-stakes exam questions, explanations, feedback, and learning loops. Your job is to make that output reliable enough that professors can inspect it, correct it, reuse it, and build confidence in the system over time.
- This is not a detached research sandbox. The work sits inside the product, close to user feedback, where reliability has to survive real usage.
What you will work on
- Evaluation loops for generated questions and explanations
- Review, retry, and failure-handling workflows
- Deterministic checks where code should beat model judgment
- Observability around bad outputs, regressions, and edge cases
- AI workflows connected to professor review and student practice
- Systems that make uncertainty visible instead of hiding it
You will be a fit if you can
- Use AI tools (for example Claude, Gemini, ChatGPT, Codex, GLM, MiniMax) as a long-term, daily part of your workflow, not occasionally.
- Have solid production experience with TypeScript, Node.js, and React.
- Independently deliver end-to-end: web UI development, backend API development, database schema design, and integration/debugging across layers.
- Have hands-on experience with at least one frontend engineering stack: Next.js, Vite, or React Router.
- Be comfortable with REST APIs, common authentication/authorization patterns, and core backend engineering fundamentals.
- Be proficient with PostgreSQL or another SQL database, including transactions, concurrency, migrations, idempotency, and soft-delete strategies.
- Be comfortable with Docker and docker-compose, and be able to set up local dev/test environments independently.
- Have real production operations experience: log-based debugging, environment/config issues, deployment failures, and baseline performance troubleshooting.
- Have real AI application engineering experience, including structured outputs, tool calling, context management, failure fallback, and regression verification.
- Learn quickly while shipping, including picking up iOS / Swift / WKWebView capabilities as needed.
- Be willing to work across layers and resolve iOS / Web / Backend / DB linkage issues.
- Preferred experience
- Experience building hybrid App / WebView / native bridge projects.
- Ability to read and modify basic Swift / SwiftUI code.
- Familiarity with practical reliability controls: caching, concurrency control, rate limiting/timeouts, graceful degradation/circuit breaking, and canary-style rollouts.
- Experience setting up observability foundations: logs, metrics, tracing, and alerting.
- Familiarity with common GCP or AWS services and basic production troubleshooting patterns (databases, object storage, WAF, load balancers).
- Experience handling production incidents such as memory pressure, CPU bottlenecks, slow queries, connection saturation, and deployment outages.
- A consistent habit of evaluating and adopting new AI workflows in real delivery contexts.
- Bonus experience
- Experience participating in SaaS system buildouts and understanding multi-tenant architecture design.
- Experience building B2B products such as productivity/collaboration tools (calendar, IM, approval, reimbursement, etc.).
- Familiarity with Grafana and Kubernetes.
- Familiarity with Vercel AI SDK.