This is a remote position.
Location: Vietnam (Remote)
Employment Type: Full-time
About the Opportunity
On behalf of our client, we are hiring a Full-Stack AI Engineer to join an innovative product team building next-generation AI-powered customer experiences.
Our client is a fast-growing technology company focused on conversational AI, intelligent automation, and customer engagement solutions. This role offers the opportunity to work on cutting-edge AI applications that combine large language models (LLMs), real-time voice interactions, and modern web technologies to create impactful user experiences at scale.
As a Full-Stack AI Engineer, you will be responsible for designing and delivering AI-powered products end-to-end - from agent orchestration and backend services to real-time user interfaces and embedded web experiences.
Key Responsibilities
Design single/multi-agent systems (OpenAI & Anthropic) with robust routing, tool calling, state, and recovery logic, primarily using Vercel AI SDK.
Build low-latency real-time voice pipelines (streaming STT/TTS, emotion-aware responses, sub-second turn-taking, and clean barge-in handling).
Operate the WebRTC-based real-time transport layer (audio rooms, agent workers, and session lifecycles).
Build Model Context Protocol (MCP) servers/clients to securely connect databases; implement agentic browser automation to execute real-world tasks.
Deliver end-to-end features using [protected info] (TypeScript) and React/Svelte, streaming agent states over AG-UI, and shipping embeddable widgets.
Tune for latency, cost, and reliability (streaming, caching, fallbacks) ; set up tracing (Langfuse, OpenTelemetry), implement prompt-injection & PII defenses, and run online A/B tests.
Requirements
Technical skills:
5+ years of production experience in [protected info] and TypeScript.
Hands-on LLM experience (tool calling, structured outputs, multi-step orchestration, context management), ideally with Vercel AI SDK.
Production experience in real-time voice/streaming media (STT/TTS, turn detection, latency tuning) and WebRTC infrastructure.
Familiarity with agent protocols (MCP, AG-UI) and agent-driven browser automation.
Strong React or Svelte skills with advanced TypeScript, and comfort building framework-agnostic, embeddable components.
Grasp of prompt engineering, agent evaluation, non-deterministic debugging, and security-first development (input validation, prompt-injection defense, secrets management, least-privilege).
Preferred Qualifications:
Bachelor's degree in Computer Science, Engineering, or equivalent
Background in conversational commerce, personalization, or consumer web CRO.
Deployment experience on AWS and Vercel serverless/edge runtimes.
Familiarity with RAG, vector databases, and retrieval pipelines.
Experience embedding JS into third-party funnels and multi-tenant environments.
LLM observability (Langfuse, OpenTelemetry) and online A/B testing.
Experience with model routing, fine-tuning, and model cost-benefit trade-offs.
Knowledge of WCAG AA accessibility for voice and conversational UIs.
Benefits
Competitive salary paid in USD via [protected info] (EOR model).
Social insurance coverage in accordance with Vietnamese regulations.
20 annual leave days plus 10 public holidays per year.
Company-provided laptop.
100% remote working environment.
Sponsored training, workshops, and industry conferences.
Clear career growth opportunities with support from both local & international teams.
How to Apply
Send your updated resume to [protected info] For more details, contact:
Ngoc Tran (Ta Hy) - Recruitment Consultant
[protected info]
Zalo/Phone number: +84 (0) 945 555 790
Thông tin chung