Lead Backend Engineer (AI Integration & Real-Time Systems) - Remote

Location: Ho Chi Minh City, Vietnam

Work Mode: Contractor | Hybrid Full-time (on-site 1–2 days/week or as required)

English Requirement: B2+ (Good communication)

Duration: 12 months (ASAP Start)

Role Overview

We are seeking a highly experienced Lead Backend Engineer to architect and implement the core backend systems for Felix AI – an innovative EdTech platform leveraging AI to deliver real-time, personalized tutoring. This role is central to building scalable, secure, low-latency systems on Google Cloud Platform, integrating advanced AI services (Gemini, Vertex AI, ADK), and supporting real-time interactions via WebSockets and audio streaming.

Key Responsibilities

● Design, implement, and scale backend services on GCP with a focus on performance, resilience, observability, and low-latency for real-time communication.

● Develop RESTful APIs using FastAPI (OpenAPI 3.0 compliant) and efficient Firestore data models.

● Integrate and orchestrate ADK-based pedagogical agents on Vertex AI, define toolkits (FunctionTools, BaseTools), and manage agent context/state.

● Implement Gemini Pro/Flash/Live APIs, including streaming logic, configuration handling, and prompt engineering for optimal AI performance.

● Architect and operate a real-time communication layer over WebSockets (Python) with secure, performant audio (PCM) streaming.

● Manage distributed session state using Redis (Memorystore).

● Collaborate with iOS engineer for API/WebSocket contracts and integration.

● Work with the Full-Stack/DevOps engineer on infrastructure, CI/CD, monitoring, and backend optimization.

● Contribute to mentoring, code reviews, documentation, and architectural decisions.

Required Qualifications

● 7–10+ years in backend engineering, with 5+ years building scalable, distributed, real-time systems.

● Expert in Python, FastAPI, WebSocket development (including TLS/WSS).

● Strong experience with GCP (Cloud Run, Firestore, Memorystore, Pub/Sub, Vertex AI).

● Proficient in integrating LLM APIs (Gemini preferred); experience with ADK or equivalent frameworks (LangChain, LangGraph).

● Hands-on experience building RAG systems (ideal in educational context).

● Familiarity with prompt engineering, distributed systems, performance tuning.

Preferred Qualifications

● Technical leadership experience in AI or EdTech projects.

● GCP certifications (Cloud Developer or Cloud Architect).

Key Performance Indicators (KPIs)

● API p95 response time < 500ms (excluding network latency)

● 99.9% backend uptime

● Complete and accurate OpenAPI documentation (>95%)

● Successful agent integration and delivery aligned with roadmap

● Load-tested scalability to target concurrency thresholds