Senior Backend Engineer: Real-Time AI Deployments

Digital Place Vision Pte Ltd; Indonesia; Sur place; IDR 360 000 000 - 720 000 000; Plein temps

Résumé du poste

Digital Place Vision Pte Ltd is hiring a Senior Backend Engineer to build and scale a real-time AI avatar platform. You will own systems transforming user voice input into GPU-rendered lip-synced video streamed in real time, covering lipsync inference, conversational orchestration, voice-cloning/TTS, and GCP deployment.

You will develop low-latency WebSocket/Protobuf services, orchestrate LLM/TTS, manage GPU inference, and ensure scalable CPU rendering paths with Dockerized Google Cloud

Qualifications

• 5+ years backend or distributed systems experience.
• Strong Python 3.10+ and asyncio expertise.
• Experience with WebSockets and low-latency streaming.
• REST APIs and gRPC/Protobuf contracts.
• Redis for session/state management.
• GCP deployment: Cloud Run, GCE, Cloud Storage.

Responsabilités

• Real-Time Streaming: build WebSocket/Protobuf services for low-latency audio/video streaming.
• AI Orchestration: coordinate LLM, TTS/voice-cloning, lipsync services.
• GPU Inference: manage PyTorch/CUDA pipelines, batching, concurrency.
• CPU Avatar Rendering: develop a CPU rendering path for cost-effective sessions.
• API & Integration: design REST, WebSocket, and Protocol Buffer APIs.
• Cloud & Reliability: deploy Docker-based services on GCP and GPUs, focusing on health checks and scaling.
• System Design: collaborate with frontend on API contracts and data flow.
• Session & Data: use Redis and GCS/Vertex AI for storage and retrieval.

Connaissances

Python 3.10+ and asyncio

5+ years experience

WebSockets

Low-latency streaming

REST APIs

gRPC/Protobuf

Redis

GPU inference

PyTorch/CUDA

Outils

Docker

Cloud Run

GCE

Cloud Storage

Artifact Registry

Redis

Description du poste

Digital Place Vision Pte Ltd is hiring a Senior Backend Engineer to build and scale a real-time AI avatar platform. You will own systems transforming user voice input into GPU-rendered lip-synced video streamed in real time, covering lipsync inference, conversational orchestration, voice-cloning/TTS, and GCP deployment.

You will develop low-latency WebSocket/Protobuf services, orchestrate LLM/TTS, manage GPU inference, and ensure scalable CPU rendering paths with Dockerized Google Cloud