Must-have profile

A proven track record of delivering LLM and RAG solutions into production for enterprise clients, ideally gained at an AI product company, consultancy, systems integrator, or industrial or energy technology firm, with hands-on experience building agentic workflows and integrating AI into existing operational systems. English required; Mandarin or Thai is a plus.

What kinds of projects you will deliver?

• LLM, conversational, and agentic AI capabilities integrated into the deployed workflows of BeyondAI's flagship operations optimization AI product for major APAC energy companies
• Conversational assistants over operational data, SOPs, maintenance records, and engineering documents
• Agentic workflows for operations decision support, such as anomaly investigation, root-cause analysis, shift reports, and recommended actions with human-in-the-loop approval
• Predictive and prescriptive analytics for asset performance, process optimization, and energy efficiency
• Enterprise knowledge platforms and RAG solutions for secure, data-sensitive environments
• Enterprise LLM projects in Taiwan and APAC, including procurement agents, quality inspection, and document AI
• Hybrid, private cloud, or on-prem AI deployments for regulated clients

Key Responsibilities

• Integrate conversational and agentic capabilities into existing operations decision-support workflows together with product and engineering teams
• Design retrieval pipelines covering document parsing, chunking, embeddings, hybrid search, reranking, and grounded answers with citations
• Build agentic workflows with tool calling, multi-step planning, guardrails, and human-in-the-loop controls
• Connect LLM applications to operational data sources such as time-series historians, maintenance systems, and enterprise databases through APIs
• Combine LLMs with classical data science and time-series ML, such as anomaly detection and forecasting, where they are the better tool
• Build evaluation frameworks with golden datasets, automated regression tests, groundedness checks, and business-level acceptance metrics
• Apply MLOps / LLMOps practices, including experiment tracking, prompt and model versioning, CI/CD, monitoring, and cost tracking
• Select and adapt LLMs, embedding models, and rerankers, and apply fine-tuning or prompt optimization where it improves accuracy or cost
• Work directly with client operations, engineering, and IT teams in workshops, UAT, and issue triage
• Translate client requirements into technical designs, tasks, and estimates with the solutions architect and delivery project manager
• Ship each milestone with release notes, documentation, and the acceptance evidence clients need to sign off
• Provide technical leadership on assigned projects by reviewing code, setting engineering standards, and guiding engineers
• Guide and review the work of engineers from delivery partners to keep quality and timelines on track
• Document repositories, environments, and runbooks so delivery never depends on a single person
• Support pre-sales with technical demos, solution input, and effort estimates
• Turn project work into reusable components and accelerators for future deployments

Required Skills and Experience

• LLM application engineering: RAG, prompt and context engineering, structured outputs, and tool calling
• Agentic AI frameworks (e.g., LangGraph, LlamaIndex, Semantic Kernel, or similar)
• LLM evaluation and observability (e.g., RAGAS, DeepEval, Langfuse, or similar)
• MLOps / LLMOps frameworks (e.g., MLflow, Kubeflow, or similar), CI/CD, and Docker
• Python, API development (FastAPI or similar), Git, and automated testing
• Data science fundamentals: statistics, time-series analysis, classical ML, and error analysis
• Vector databases and search (e.g., Milvus, pgvector, OpenSearch) and SQL
• Cloud, on-prem, and hybrid deployment of AI applications
• Clear technical communication in English with clients and cross-functional teams
• 3+ years of experience in data science, ML engineering, or applied AI
• 2+ years of experience delivering LLM/GenAI solutions, including RAG and agents, to production for enterprise users
• Proven experience in client-facing project delivery through to client acceptance
• Experience integrating AI into existing enterprise applications or operational workflows
• Experience applying MLOps / LLMOps practices in production environments

Preferred Qualifications

• Experience in energy, oil and gas, refining, power, utilities, or industrial operations
• Experience with industrial data such as process historians, SCADA/IoT data, and maintenance management systems
• Experience with on-prem AI hardware and inference optimization, such as GPU sizing, model serving (e.g., vLLM), and quantization
• Experience with Kubernetes
• Experience as a technical lead or mentor, including for distributed or partner teams
• Mandarin or Thai language skills
• Master's or PhD in Computer Science, Data Science, Statistics, Engineering, or a related field

‍

‍