Back to search

AI /LLM Application Engineer

Build and ship production GenAI and agentic systems, design RAG pipelines, integrate models, and ensure secure, governed, observable AI solutions.

  • Kuala Lumpur, Kuala Lumpur, Malaysia
  • Full-time
  • On-site
  • Posted Nov 12, 2026
INDUSTRYSoftware & InternetSaaS & Enterprise Software

Responsibilities

  • Build and ship GenAI applications and agent-first systems
  • Design and implement RAG pipelines
  • Integrate frontier and open-weight models
  • Implement security and governance controls
  • Document architectures and produce evidence for governance

Skills

PythonAPI developmentmicroservicescloud-native engineeringRAGvector databasesembedding modelsretrieval strategiesagentic frameworksmanaged AI platformsLLM fundamentalspromptingtool useevaluationlatencycost trade-offscontext window managementconsultingclient-facingagent evaluationguardrailssecurity patternsMLOpsLLMOpsmodel registriesexperiment trackingCI/CDmonitoringData engineeringDatabricksSnowflakelakehouse patternsSparkSQLknowledge graphsLangGraphCrewAIAutoGenLangChainSemantic KernelAzure OpenAIAzure AI FoundryAWS BedrockGCP Vertex AIPineconeWeaviatepgvectorAzure AI Search

Full description

US MNC scaling its Data & AI practice in Singapore is now adding a hands-on GenAI Engineer to the core build team. This is a builder role. You will work directly under the Lead AI Architect to take enterprise AI and agentic use-cases from design to production. If the Architect defines what and why, you own how it gets built, deployed, governed, and operated. You will be part of a small, senior team in Singapore [AI Architect, AI Engineers, Data Engineer, Data Scientists] and work with global Data & AI and Controls teams, plus client engineering and architecture teams. What You'll Build 1. Build Production GenAI & Agentic Systems Build and ship GenAI applications and agent-first systems - from POC to production. This includes multi-agent workflows, tool-calling agents, orchestration using LangGraph / CrewAI / AutoGen / Microsoft Agent Framework, MCP servers, model gateways, and integration with enterprise systems via APIs and events. 2. Own RAG & Knowledge Layer Design and implement RAG pipelines - ingestion, chunking, embedding, vector stores, hybrid search, re-ranking, knowledge graphs and semantic layers. Optimize for accuracy, latency, cost, and grounding. Build evaluation harnesses for retrieval quality, hallucination, and answer relevance. 3. Model Integration & Platform Engineering Integrate frontier and open-weight models - Claude, GPT, Gemini, Llama, Gemma, Phi, Mistral etc. - plus APAC / sovereign models where needed - Qwen, SEA-LION, etc. Work across Azure AI Foundry / AOAI, Bedrock, Vertex AI and handle prompt engineering, structured output, function calling, context management, and guardrails. Manage model routing, fallbacks and cost controls. 4. Ship it Right - Secure, Governed, Observable Build with security and controls from day zero - prompt injection defense, tool authorization, least-privilege identity, DLP, human approval gates, audit logging. Implement observability, evals, monitoring, and CI/CD for AI systems. Document architectures and produce evidence for governance / audit. What We're Looking For Experience in software engineering / data / ML engineering with at least 2+ years hands-on shipping production GenAI systems. Strong Python, with experience in API development, microservices, and cloud-native engineering. Proven experience building RAG - vector DBs [Pinecone, Weaviate, pgvector, Azure AI Search etc.], embedding models, and retrieval strategies. Hands-on with at least one agentic framework - LangGraph, CrewAI, AutoGen, LangChain, Semantic Kernel or similar. Experience with managed AI platforms - Azure OpenAI / Foundry, AWS Bedrock, GCP Vertex AI. Understanding of LLM fundamentals - prompting, tool use, evaluation, latency / cost trade-offs, and context window management. Comfortable working in consulting / client-facing environment - you can translate requirements and demo working software to technical stakeholders. Strong Advantage If You Have: Experience with agent evaluation, guardrails and security patterns for agentic AI. Knowledge of MLOps / LLMOps - model registries, experiment tracking, CI/CD, monitoring. Data engineering - Databricks / Snowflake, lakehouse patterns, Spark / SQL, knowledge graphs.

AI /LLM Application Engineer at Newbridge Inc | Referso Jobs