Staff AI Engineer

Wati
Shenzhen
Workplace: OnsiteFull timeFunction: Data Science & Machine LearningExperience: 5+ yearsSkills: ["Autonomous","Builder mindset","Systems thinking","Technical leadership","Product-minded"]

Own LLM orchestration, RAG, and agent infrastructure for a customer engagement platform processing 4B+ messages annually across 100+ countries. Lead architecture, deployment, and optimization of multi-provider inference routing (OpenAI, Gemini, and others), scalable RAG pipelines, and multi-agent workflows with tool-calling. Drive voice and multimodal AI across text/voice channels, build production AI lifecycles (data pipelines, fine-tuning orchestration, versioning), and improve quality via evaluation and benchmarking.

Loading

Loading job details...

Preparing the role view and application actions.

FursaFursa
Wati
Wati
4 months ago

Staff AI Engineer

✓ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI
Last checked: 1 hour agoStatus: Live

Job Summary

Own LLM orchestration, RAG, and agent infrastructure for a customer engagement platform processing 4B+ messages annually across 100+ countries. Lead architecture, deployment, and optimization of multi-provider inference routing (OpenAI, Gemini, and others), scalable RAG pipelines, and multi-agent workflows with tool-calling. Drive voice and multimodal AI across text/voice channels, build production AI lifecycles (data pipelines, fine-tuning orchestration, versioning), and improve quality via evaluation and benchmarking.
Location: Shenzhen
Workplace: Onsite
Employment Type: Full time
Job Function: Data Science & Machine Learning
Seniority: Mid level

Key Responsibilities

  • •Architect and lead an AI production stack, including multi-provider LLM gateway optimization, token budget management, and low-latency inference routing.
  • •Design and implement scalable RAG systems and multi-step agent workflows, including tool-calling infrastructure (MCP) for reliable customer interactions.
  • •Lead voice AI evolution (WebRTC/realtime) and cross-channel agent coordination across text, voice, and connected messaging platforms.
  • •Own the engineering-to-AI production lifecycle: data collection/cleaning, fine-tuning orchestration, and model versioning pipelines.
  • •Continuously optimize performance and cost (latency, token budgets, caching) and build evaluation/benchmarking infrastructure to assess AI quality in production.

Key Requirements

  • •5+ years of professional experience in backend or infrastructure engineering, with mastery of at least one high-performance language (Go, Rust, or C++) and deep proficiency in Python.
  • •Proven track record taking LLM/NLP models from experiments to high-traffic production, including multi-provider orchestration and model drift management.
  • •Strong experience building data pipelines for AI workloads, including document processing, embedding generation, and vector search.
  • •Experience with vector databases (e.g., Qdrant, Milvus, Pinecone) and RAG architecture patterns.
  • •Familiarity with agentic frameworks/tool-calling protocols (MCP, function calling) and real-time voice/audio AI pipelines (WebRTC, LiveKit or similar).
Experience:5+ years
Skills:AutonomousBuilder mindsetSystems thinkingTechnical leadershipProduct-minded
Tech Stack:GoRustC++PythonOpenAIGeminiRAGMulti-provider orchestrationMCPFunction callingWebRTCLiveKitGCPAWSDockerKubernetesQdrantMilvusPineconeVector databases

Company Brief

Wati
Provides a WhatsApp-based customer engagement platform that helps businesses manage sales, support, and marketing conversations in one shared inbox. The product includes automation, broadcasts, chatbots, and team collaboration tools for messaging-led workflows.
Industry: SaaS
Company Size: Medium (51 to 250 employees)
Growth: Growth Stage Startup
Headquarters: Hong Kong, Hong Kong
Founded: 2018
WebsiteLinkedIn