New opportunity

Senior Data Scientist (Generative AI)

Razer · Singapore

About this role

The Senior Data Scientist (Generative AI) will design and develop GenAI-driven solutions for interactive systems and developer-facing tools at Razer. This role involves building scalable, production-ready AI services and collaborating with cross-functional teams to integrate AI into gaming ecosystems. Responsibilities include: Designing and maintaining GenAI solutions for developer tools; building RAG-based systems including retrieval pipelines and prompt orchestration; developing agentic workflows for reasoning and planning; managing production pipelines for GenAI microservices; fine-tuning LLMs for domain-specific tasks; and optimizing services for latency, scalability, and cost. Qualifications: Master’s or PhD in a relevant field (Computer Science, AI, Machine Learning, etc.) and at least 2 years of applied experience in machine learning or Generative AI. Required skills include proficiency in Python, deep learning frameworks (PyTorch or TensorFlow), and experience with RAG pipelines and cloud-based ML workflows. Preferred qualifications include experience with multi-GPU training, distributed experiments, and AWS SageMaker. The role is based in the Singapore office with occasional travel.

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

Skills for this role

PythonGenerative AILarge Language ModelsRAGAgentic workflowsPyTorchTensorFlowPrompt engineeringMachine learningCloud-based AI platformsAWS SageMakerMulti-GPU trainingDistributed experimentsModel fine-tuning

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

Production-Ready GenAI Development
The role requires moving beyond experimentation to building scalable, production-ready AI services. You must demonstrate the ability to manage the full lifecycle of GenAI microservices, including deployment, monitoring, and maintenance within cloud-based infrastructures.
RAG and Agentic Workflow Implementation
A core priority is the design and extension of RAG-based systems and agentic workflows. You will be expected to handle complex tasks such as retrieval pipeline optimization, prompt orchestration, and developing agents capable of reasoning and planning.
Cross-Functional Collaboration
Success depends on working closely with software, data, and game developers. You must be able to define data schemas, coordinate with engineering teams on production pipelines, and communicate technical AI concepts to non-technical stakeholders.
Model Optimization and Fine-Tuning
You are responsible for fine-tuning LLMs for domain-specific tasks while ensuring services are optimized for latency, scalability, and cost. This requires a rigorous approach to evaluating performance using task-specific metrics and benchmarks.

Likely challenges to think through

Balancing Latency and Complexity in Gaming
Interactive gaming systems often require extremely low latency. The challenge lies in implementing sophisticated agentic workflows or RAG systems without introducing performance bottlenecks. Prepare by discussing strategies for model quantization, caching, or asynchronous processing in production.
Managing Data Pipelines for Diverse Inputs
You will need to define validation strategies for both structured and unstructured data. The difficulty is ensuring data quality for GenAI systems that may ingest varied gaming-related data. Prepare by highlighting your experience in building robust data ingestion and cleaning pipelines.
Scaling Infrastructure for AI Services
The role involves managing production pipelines and potentially multi-GPU training. The challenge is maintaining cost efficiency while scaling. Prepare by discussing your experience with cloud-based ML platforms like AWS SageMaker and how you approach resource allocation for distributed experiments.

Practice your answers

Technical practice

Designing RAG Retrieval Pipelines

What to coverDiscuss how you would optimize retrieval accuracy in a RAG system. Address strategies for indexing, chunking, and re-ranking, and explain how you would evaluate the retrieval quality using specific metrics.

Agentic Workflow Architecture

What to coverExplain your approach to designing agentic workflows that require reasoning and tool-based execution. Discuss how you handle state management, error recovery, and the trade-offs between autonomous agent behavior and deterministic control.

LLM Fine-Tuning for Domain Specificity

What to coverDescribe the process of fine-tuning an LLM for a specific domain. Focus on data preparation, selection of fine-tuning techniques (e.g., LoRA, QLoRA), and how you measure performance improvements against a baseline.

Productionizing GenAI Microservices

What to coverHow do you ensure GenAI services are production-ready? Cover topics such as latency optimization, cost management, monitoring for model drift, and ensuring reliability in a high-traffic environment.

Distributed Training and Multi-GPU Scaling

What to coverDiscuss the challenges of scaling model training across multiple GPUs. Explain how you manage data parallelism, model parallelism, and synchronization overhead to optimize training time.

Prompt Orchestration Strategies

What to coverExplain how you manage complex prompt chains or orchestration in a production system. Discuss versioning, testing prompts for consistency, and how you handle prompt injection or safety concerns.

Data Ingestion and Validation for AI

What to coverDescribe how you collaborate with data engineers to build robust ingestion pipelines. Focus on how you define schemas and implement validation strategies to ensure the quality of both structured and unstructured data used by GenAI models.

Evaluating GenAI System Performance

What to coverBeyond standard accuracy metrics, how do you evaluate the performance of a generative system? Discuss the use of qualitative analysis, human-in-the-loop evaluation, and automated benchmarks for specific tasks.

Behavioural practice

Cross-Functional Collaboration

What to coverDescribe a time you worked with non-technical stakeholders or different engineering teams to deliver an AI project. Focus on how you communicated technical constraints and aligned the project goals with their requirements.

Handling Ethical and Regulatory Challenges

What to coverProvide an example of how you have addressed ethical or regulatory implications in an AI project. Explain your decision-making process and how you balanced innovation with safety or compliance requirements.

Problem-Solving Under Technical Constraints

What to coverTell me about a time you had to optimize a model or pipeline for strict latency or cost constraints. Describe the specific actions you took to identify the bottleneck and the final result of your optimizations.

Adapting to Emerging Technologies

What to coverThe GenAI field evolves rapidly. Describe a situation where you had to quickly learn and implement a new technique or framework to solve a project challenge. How did you validate its effectiveness?

Sharpen your knowledge

Topics to prepare

Agentic Workflows and Reasoning
The role requires developing agentic workflows for reasoning, planning, and tool-based execution. Exercise: Design a conceptual architecture for an AI agent that can interact with a game engine API to perform automated tasks, focusing on how the agent handles planning steps and tool selection.
RAG System Architecture
The position involves building and extending RAG-based systems, including retrieval pipelines and indexing strategies. Exercise: Review different vector database indexing strategies and retrieval techniques (e.g., hybrid search, reranking) to optimize for latency and accuracy in a production environment.
Production ML Pipelines
The role requires managing production pipelines for GenAI microservices and optimizing for scalability and cost. Exercise: Outline a CI/CD pipeline for a GenAI service, identifying key stages for automated testing, model monitoring, and performance benchmarking.

Skills to revise

Deep Learning Frameworks (PyTorch/TensorFlow)
Hands-on experience with PyTorch or TensorFlow is required for model development. Exercise: Implement a custom training loop or a model fine-tuning script in your preferred framework to demonstrate proficiency in handling data loaders and model optimization.
Cloud-based ML Workflows (AWS SageMaker)
Experience with cloud platforms like AWS SageMaker is preferred for deployment and training. Exercise: Create a mock configuration for deploying a model endpoint on AWS SageMaker, focusing on setting up auto-scaling policies and monitoring metrics.
Model Fine-tuning and Optimization
The role involves fine-tuning LLMs and optimizing for latency and cost. Exercise: Practice fine-tuning a small open-source LLM on a domain-specific dataset, then measure the impact of quantization or pruning on inference latency.

Your preparation checklist

Build a RAG-based microservice project
Develop a functional RAG pipeline using Python and PyTorch or TensorFlow. Ensure the project includes clear retrieval strategies, prompt orchestration, and a documented evaluation of performance using task-specific metrics to demonstrate your ability to build production-ready AI services.
Prepare an agentic workflow demonstration
Create a small-scale agentic system that utilizes tool-calling or planning-based workflows. Be prepared to explain how you managed reasoning patterns and how you would scale such a system for interactive gaming tools.
Document your experience with cloud-based ML pipelines
Prepare a summary of your experience with AWS SageMaker or similar platforms, focusing on how you have managed production pipelines, model fine-tuning, and distributed training experiments in previous roles.
Formulate questions regarding Razer's AI ecosystem
Prepare thoughtful questions about how GenAI is currently integrated into Razer's developer tools or gaming hardware, and inquire about the team's approach to balancing model latency, cost, and scalability in their production environments.
Review your technical communication skills
Practice explaining complex GenAI concepts, such as fine-tuning strategies or RAG indexing, to both technical peers and non-technical stakeholders, as this role requires cross-functional collaboration with game developers and software engineers.

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…