Senior AI Engineer

Role Summary
We are seeking a highly skilled Senior AI Engineer with at least 3 years of experience to lead the integration of Generative AI into our enterprise ecosystem. You will design, build, and optimize AI-powered platforms that enable our teams and customers to interact with internal documentation and data using natural language.
Key Responsibilities
AI Architecture & Implementation: Design and implement system architectures for AI knowledge retrieval using Retrieval-Augmented Generation (RAG) pipelines.
LLM Integration: Integrate multiple LLM providers including OpenAI (GPT), Anthropic (Claude), and Google Gemini.
Vector Database Management: Build and maintain semantic search capabilities using vector databases such as Pinecone, FAISS, or Weaviate.
Agentic Workflows: Develop AI Agent workflows for automated tasks including document summarization and intelligent knowledge query automation.
Prompt Engineering: Apply advanced prompt engineering techniques (Zero-shot, Few-shot, Chain-of-Thought) to optimize model responses and mitigate hallucinations.
Platform Integration: Build scalable backend APIs using [protected info], Python, or Spring Boot to integrate AI services with existing web applications and internal tools.
Custom Model Development: Possibility of creating new AI models aligned with clients' specific business requirements.
Own the full lifecycle of LLM initiatives: opportunity sizing, experiment design, dataset curation and labeling, offline evaluations, A/B tests, launch, and post-launch monitoring.
Build LLM systems such as RAG, function/tool use, and lightweight fine-tuning (LoRA/PEFT) for domain-specific tasks.
Design rigorous evaluation frameworks for generative systems - covering truthfulness, task success, hallucination rate, safety guardrails, latency, and cost per output token - and maintain dashboards and alerts.
Productionize models and pipelines with robust data preprocessing, vector indexing, caching, fallback policies, and prompt/version management.
Optimize for quality and cost via model selection and routing, context window strategy, compression, and inference tuning.
Mentor DS/ML peers; contribute to best practices, documentation, and design reviews.
Shift Requirements: The role is fully remote / in office, requiring a minimum of 8 hours of work per day, of which 4 hours must overlap with 8:00 PM to 12:00 AM Vietnam time zone