Backend Engineer_AI Gateway (Korean Required)

We are building an AI Gateway-a high-performance intermediary that routes, inspects, and governs traffic between enterprise clients and LLM/AI service providers. In this role you will own the entire full-stack lifecycle of the gateway, from core Python API service design and implementation to containerization, deployment, monitoring, and day-to-day operations. While you can consult network and security specialists for architecture reviews, threat modeling, and policy definition, the application code, testing, CI/CD pipeline, and production operations will be developed and maintained solely by you. You’ll have full technical autonomy to design, build, and scale a robust, production-grade system.

• Experience: Around 5 years of professional backend development experience, with a solid track record of building and operating production-grade services.
• Async Python & Core Web Frameworks (3+ yrs): Deep understanding of asyncio and hands-on production experience with FastAPI (Pydantic, dependency injection, OpenAPI).

• Network Protocol & Streaming Fundamentals: o Strong grasp of HTTP/1.1, HTTP/2, and SSE (Server-Sent Events) / Chunked transfer encoding essential for real-time LLM inference streaming. o Practical understanding of the TCP/IP lifecycle, connection pooling, keep-alive, and managing back-pressure.

• Database & Async Messaging: o Proficiency in PostgreSQL (query tuning, connection pooling via PgBouncer). o Experience with message queues or event streams (e.g., Redis Streams, RabbitMQ, or Kafka) for decoupled task processing.

• Containerization & CI/CD: o Docker, Kubernetes (basic workload management, deployment, HPA), and setting up CI/CD pipelines (GitHub Actions / GitLab CI). Work experience desired

• Collaborative Security & Infra Awareness: Willingness to work with internal security/network teams on NGINX/Envoy reverse-proxy setup, mTLS, or IAM/SSO (OIDC/SAML) integrations.
• AI/LLM Integration: Experience handling LLM API payloads, managing rate limits, or implementing simple proxy/gateway patterns for AI services.
• Infrastructure as Code: Basic familiarity with Terraform for multi-env infrastructure management.