New opportunity

AI Engineer

Surbana Jurong · Singapore

About this role

The AI Engineer will join the Central AI-Core Studio & Delivery team at Surbana Jurong to drive the design, development, and production delivery of AI-enabled applications for core business lines. This role involves managing the full software development lifecycle, from initial requirements and architecture to testing, deployment, and post-launch iteration. Key responsibilities include building end-to-end AI-powered applications, integrating LLM APIs and vector databases, implementing retrieval-augmented generation, and applying agentic orchestration. The engineer will also define evaluation criteria for AI models, manage CI/CD pipelines, ensure secure cloud deployment, and contribute reusable components to the company's AI platform. Collaboration with business teams and AI Product Managers is essential in this forward-deployed model. Required qualifications include a proven track record of shipping production software, strong Python proficiency, and hands-on experience with LLM APIs and retrieval systems. Candidates should possess a solid understanding of software architecture, testing practices, and production operations. Advantageous qualifications include experience with agentic orchestration frameworks (LangChain, LangGraph, AutoGen), frontend development (React, TypeScript), and familiarity with the Architecture, Engineering and Construction (AEC) industry or enterprise workflow domains, as well as cloud environments like AWS, Azure, or GCP.

For full details about this role, click “Apply on company site” to view the employer’s original posting.

Skills for this role

PythonREST APIsCloud deploymentContainerisationLLM APIsOpenAIAnthropicRetrieval-Augmented GenerationVector databasesSoftware architectureCI/CDObservabilityLangChainLangGraphAutoGenReactTypeScriptAWSAzureGCP

Checking your profile…

Preparation suggestions, not the employer’s actual interview questions. Always check the original posting for current requirements.

Understand the role

What matters most

End-to-End Production Delivery
The role requires owning the full software development lifecycle for AI applications. You must demonstrate the ability to take a project from initial requirements and architecture through to deployment, testing, and post-launch iteration, ensuring the software is reliable under real operating conditions.
LLM Integration and RAG Implementation
A core priority is building functional AI applications using LLM APIs (OpenAI, Anthropic) and vector databases. You must be proficient in implementing retrieval-augmented generation (RAG) and applying agentic orchestration to solve complex business problems.
Model Evaluation and Performance Iteration
You are expected to define clear evaluation criteria for AI models and retrieval systems. This involves using production performance data to iteratively improve the accuracy and effectiveness of the AI solutions deployed for business lines.
Stakeholder Collaboration in a Forward-Deployed Model
You will work directly with business teams and AI Product Managers. Success depends on your ability to communicate technical progress, risks, and architectural trade-offs to both technical and non-technical stakeholders effectively.

Likely challenges to think through

Balancing Rapid Prototyping with Production Stability
In a forward-deployed model, there is often pressure to deliver AI features quickly. The challenge is maintaining robust CI/CD pipelines, observability, and secure cloud deployment standards while iterating rapidly. Prepare by discussing your experience with automated testing and maintaining production-grade software.
Navigating Complex Domain-Specific Workflows
Surbana Jurong operates in the AEC (Architecture, Engineering, and Construction) industry. You may face difficulty translating abstract AI capabilities into practical solutions for specialized enterprise workflows. Prepare by researching how AI is currently applied in infrastructure or engineering sectors to demonstrate domain awareness.
Managing Non-Deterministic AI Behavior
Integrating LLMs into production introduces non-deterministic outputs that can be difficult to debug. You will need to establish rigorous evaluation frameworks to manage this. Prepare by explaining your approach to monitoring model drift and implementing guardrails for LLM-based applications.

Practice your answers

Technical practice

Designing RAG Pipelines for Production

What to coverExplain your approach to building a robust Retrieval-Augmented Generation pipeline. Focus on how you handle document chunking, embedding strategies, and the trade-offs between different vector database indexing methods to ensure low-latency retrieval.

Evaluating LLM Application Performance

What to coverDescribe how you define and implement evaluation criteria for AI models in a production environment. Discuss the use of automated metrics versus human-in-the-loop feedback and how you iterate on performance based on real-world data.

Agentic Orchestration Patterns

What to coverDiscuss your experience with frameworks like LangChain or LangGraph for agentic workflows. Explain how you manage state, handle tool-use loops, and prevent infinite execution cycles in complex agentic systems.

CI/CD and Deployment for AI Services

What to coverDetail your strategy for managing CI/CD pipelines for AI-enabled applications. Address how you handle containerisation, versioning of model artifacts, and ensuring secure, scalable deployments on cloud platforms like AWS, Azure, or GCP.

Observability in LLM Systems

What to coverExplain how you monitor production LLM applications to detect drift, latency issues, or unexpected model behavior. Suggest specific tools or logging strategies you use to maintain visibility into the full request-response lifecycle.

Software Architecture for AI Integration

What to coverHow do you integrate LLM APIs into existing backend architectures without creating tight coupling? Discuss the design patterns you use to ensure the system remains modular and testable as requirements evolve.

Handling Production Failures

What to coverDescribe a scenario where an AI-powered feature failed in production. Explain your debugging process, how you identified the root cause (e.g., prompt issues, retrieval errors, or API instability), and the steps taken to prevent recurrence.

Behavioural practice

Collaborating with Non-Technical Stakeholders

What to coverProvide an example of a time you had to explain a complex AI technical risk or progress to a non-technical Product Manager or business stakeholder. How did you ensure they understood the implications and aligned on the next steps?

Managing Conflicting Requirements

What to coverDescribe a situation where you had to balance the need for rapid AI model iteration with the requirement for stable, production-grade software. How did you negotiate these priorities with your team?

Driving Technical Standards

What to coverGive an example of how you contributed to reusable components or engineering standards within a team. What was the impact of these contributions on the team's overall development velocity?

Adapting to Domain-Specific Challenges

What to coverIf you have worked in a complex industry like AEC or enterprise workflows, describe a time you had to learn a new domain to build an effective technical solution. How did you bridge the gap between domain knowledge and technical implementation?

Sharpen your knowledge

Topics to prepare

Retrieval-Augmented Generation (RAG) Architecture
The role requires designing and building retrieval pipelines. Exercise: Draft an architectural diagram for a RAG system that handles document ingestion, vector database storage, and context-aware retrieval. Focus on how you would define evaluation criteria for retrieval accuracy and model performance.
Agentic Orchestration Concepts
The role involves applying agentic orchestration. Exercise: Review the conceptual differences between simple LLM chains and agentic workflows (e.g., using LangGraph or AutoGen). Prepare to explain how you would manage state and tool-use loops in a production environment.
Production Software Lifecycle
The role emphasizes the full software development lifecycle. Exercise: Outline a strategy for maintaining an AI application post-launch, specifically focusing on how you would implement observability to monitor model drift and system performance in a cloud environment.

Skills to revise

Python for Production AI
The role requires strong Python proficiency for building backend APIs and AI integrations. Exercise: Refactor a script that interacts with an LLM API to include robust error handling, logging, and asynchronous execution to ensure it is production-ready.
CI/CD and Containerisation
The role requires owning CI/CD pipelines and containerisation. Exercise: Create a Dockerfile for a Python-based AI application and write a basic CI/CD configuration file (e.g., GitHub Actions or GitLab CI) that automates testing and deployment to a cloud provider.
Vector Database Integration
Hands-on experience with vector databases is required. Exercise: Build a small prototype that indexes a set of documents into a vector database and performs a semantic search, ensuring you can explain the trade-offs of your chosen embedding model and similarity metric.

Your preparation checklist

Prepare a production-grade AI project walkthrough
Select a past project where you managed the full lifecycle of an AI application. Be prepared to discuss your architectural choices, how you handled LLM API integration, the specific vector database used, and how you implemented RAG or agentic workflows to solve a real-world problem.
Document your approach to AI model evaluation
Create a brief summary of how you define success metrics for AI applications. Explain your process for evaluating retrieval quality and model performance, and describe how you iterate on these metrics using production data to improve system reliability.
Review your experience with CI/CD and cloud operations
Prepare to explain your experience with containerization and cloud deployment (AWS, Azure, or GCP). Be ready to discuss how you set up CI/CD pipelines, ensure secure deployments, and implement observability tools to monitor application health in production.
Draft technical questions for the interviewers
Prepare thoughtful questions regarding the Central AI-Core Studio's current tech stack, how they balance building reusable platform components versus custom business-line solutions, and the specific challenges they face when deploying AI in the AEC or infrastructure domains.
Reflect on AEC industry applications
Research the Architecture, Engineering, and Construction (AEC) sector to understand common enterprise workflows. Prepare to discuss how your AI engineering skills could specifically address challenges in infrastructure project lifecycles, planning, or facility management.

Availability last checked: 18 Oct 2026. Status can change after a check; confirm on the employer’s website. Report an issue · Our checking policy

KEEP EXPLORING

Similar AI jobs

Related skills and specializations in Singapore. Matched to this role, not your profile.

Explore more jobs

Finding similar opportunities…