Product Engineer

About Praxis

Praxis is an AI-native operating company that believes in every person's capacity to grow and builds better systems for learning. We create and operate ELA, Epic Prep, Rubric, and Playfit. In this role, you would join the Praxis HQ team. We do not offer a hybrid work arrangement.

What You’ll Do

• Observe students, parents, teachers, coaches, and operators across every service, then define product requirements, success metrics, and the technical approach with them.
• Design, build, and operate the web frontend, APIs, data models, authentication and authorization, and deployment infrastructure.
• Integrate LLMs, agents, retrieval, and evaluation pipelines while managing accuracy, cost, latency, and safety through eval sets and telemetry.
• Use AI coding agents as part of the daily development environment while retaining full responsibility through tests, review, and security validation.
• Build observability, testing, CI/CD, incident response, and data-protection systems, then turn failures into prevention automation and documentation.
• Prioritize from user interviews and product metrics, ship in small increments, and learn quickly from production.
• Turn recurring support and field-operations problems into product features, internal tools, and automations.
• Work across every Praxis service and the design, education, coaching, and operations teams, owning the result from idea through operating outcomes.

What We’re Looking For

• Bachelor’s degree in computer science, software engineering, data, or a related technical field
• At least four years of experience building commercial web products, including production ownership of both frontend and backend
• Strong fundamentals in TypeScript/JavaScript, React or a comparable framework, Node.js or Python, SQL/PostgreSQL, cloud systems, and containers
• A case where you owned requirements, architecture, implementation, deployment, monitoring, and user-feedback iteration end to end
• Hands-on production experience across more than one of these areas: LLM APIs, RAG/retrieval, agent workflows, structured outputs, and evals
• The judgment to move faster with AI coding tools without lowering standards for tests, review, security, or privacy
• The agency to propose alternatives in ambiguity, meet users directly, and optimize for outcomes rather than technology
• A habit of turning personal know-how into code, documentation, automation, and dashboards that compound team execution

Nice to Have

• Experience with education technology, B2B SaaS, multi-tenant authorization, or payments
• Experience building LLM evaluation, observability, prompt-versioning, or AI-safety systems
• Experience with infrastructure as code, Kubernetes/serverless, data pipelines, or MLOps
• Open-source work, technical writing, or a portfolio with real service metrics
• The ability to conduct technical documentation and user interviews in Korean and English

Hiring Process

1. Application and representative project/code review
2. Full-stack systems and product-judgment interview
3. AI pair-building and code-review session
4. Field problem-solving and values interview
5. Reference checks
6. Compensation discussion informed by industry standards