Middle/Senior AI Engineer - Fluent Eng

1 - Project Overview:We are building an enterprise-grade Agentic AI platform powered by LLMs, RAG, and custom orchestration. Unlike typical GenAI teams, we do not rely on a single framework. We design systems from first principles and build production-ready AI systems, not demos.- Architect and develop specialized, enterprise-grade AI Agents capable of reasoning, tool-calling, automation, and long-term memory.- Implement intelligent workflows using RAG, embeddings, ontology-based reasoning, and AI memory structures 2 - Job Description:Primary Skills (Must-have)LLM & RAG Systems- Agentic workflows: planning, reflection, self‑improvement loops- Build production RAG pipelines and LLM applications- End-to-end pipelines: data → embeddings → retrieval → LLM reasoning → evaluation → deployment- Experience with embeddings and vector DBs (Qdrant, FAISS, Pinecone, pgvector)- Vector search & retrieval optimizationBackend Engineering- Strong Python- API development (FastAPI/Flask)- Scalable backend systems, async, cachingSystem Design- End-to-end AI system design- Agent memory custom design- Knowledge base solution design- Performance tuning (retrieval, prompts, caching)Secondary Skills- Framework exposure: LangChain, LlamaIndex.- Cloud: Hand-on experience at least on cloud service GCP/AWS/Azure- MLOps: Docker, Kubernetes, MLflow- Google Vertex AI /Azure AI Foundry/ Amazon Bedrock- Running self-hosted LLMs- Fine-tuning LLMs directly on cloud GPUs- Automated evaluation pipelines: grounding, hallucination detection, drift analysis- Deploy AI systems with monitoring, observability, and safety guardrails 3 - Qualifications:- 3+ years in Applied AI / ML Engineering with production systems- Strong Python skills for AI pipelines and automation- Experience building LLM-based products, RAG systems, or Agentic workflows- Knowledge of vector DBs, embeddings, retrieval optimization- Familiarity with MLOps tools (MLflow, Helm, Kubernetes)- Experience with LLMs (OpenAI, HuggingFace, Ollama)- Ability to operate in a fast-paced engineering team 4 - Others/Benefit:- Work with cutting-edge AI models and deep-tech innovation- Build specialized agentic systems for enterprise workflows- Experiment with fine-tuning, self-hosting, embeddings, multimodal models- Fast-growing AI team with significant career growth- Collaborate with engineers from Amazon, Microsoft, Google- Direct impact: systems you build go into production“FPT Care” health insurance provided by PJICO (exclusive for FPT employees).Annual Summer Vacation following company policy (starting from May each year).Salary review once per year.International, dynamic, and friendly working environment.Annual leave and working conditions in compliance with Vietnam Labor Law.Sponsorship for training programs and international certification exams.Preferential loan interest support policy for FPT Software employees.Shuttle bus service for employees.