AI Engineer

HMD is a company specializing in online product sales in the US and providing operational and support services for Vietnamese business owners to introduce their products to the world market. With 10 years of experience in e-commerce in the US, we possess the strengths and insights necessary to effectively and optimally manage, brand, and drive success for Vietnamese products.View more: https://behmd.comJob SummaryThis role is ideal for someone passionate about large language models (LLMs), eager to contribute to real-world applications, and capable of supporting the full lifecycle from development to deployment. LLM Engineer will collaborate with AI Product Managers, ML Engineers, and other stakeholders to bring cutting-edge natural language solutions to production.Responsibilities1. Model Development & OptimizationFine-tune open-source or foundation LLMs for specific use casesBuild and manage prompt templates and chains using frameworks like LangChainPrepare and preprocess data for model training and evaluationContinuously improve model performance, latency, and accuracy through iteration2. Deployment & InfrastructureIntegrate LLMs with retrieval systems (e.g., RAG pipelines)Develop APIs or middleware to serve LLM outputs to applicationsMonitor and optimize inference performance and resource usage3. Evaluation & MonitoringDesign benchmark test cases and golden datasets for model evaluationTrack performance metrics such as accuracy, hallucination rate, response timeImplement feedback loops and assist in A/B testing for model iteration4. AI Safety & ComplianceSupport efforts in bias detection, content moderation, and ethical AI usageEnsure outputs adhere to quality, privacy, and regulatory standards5. Collaboration & ResearchWork cross-functionally with DA, DE, and AI PM to identify NLP-driven use casesCollaborate with other business departments to identify opportunities where LLMs can optimize workflows and enhance operational efficiencyStay updated with emerging trends in generative AI, transformers, and foundation modelsRequirementsBachelor's or Master's degree in computer science, artificial intelligence, or a related field.Strong understanding of transformer architectures and LLM internals (e.g., attention mechanism, positional encoding, tokenization)Familiarity with RAG architecture and document retrieval pipelinesKnowledge of AI/ML deployment best practices, including scalability and inference optimizationAwareness of ethical AI practices, hallucination mitigation, and responsible AI use1–3 years of experience working with NLP or LLM-based systems in production or researchProficient in Python; familiarity with PyTorch, Hugging Face, LangChain, and REST APIsSolid grasp of prompt engineering, embeddings, and vector databases (e.g., FAISS, Pinecone)Understanding of training/fine-tuning workflows and performance tuningExperience with GenAI infrastructure tools (e.g., MLflow, Kubeflow, Airflow)Experience with prompt management strategies, versioning, and testingFamiliarity with safety frameworks, ISO 27001/SOC2 standards, or AI governance principlesNice to HaveExperience in e-commerce is a strong advantage.BenefitsCompetitive income compared to the marketEnjoy the Company's comprehensive health programHolidays: 12 days of annual leave & statutory holidaysBonus: Holiday, Tet, sales,...Giving gifts on: March 8 (women), April 6 (man), birthday,...Allowances: marriage, bereavement, sickness, maternityParticipate in annual activities: Year-end, New Year, Teambuilding,...Working time: 8am - 5pm, Monday - Friday