Software Engineer, AI Infrastructure & Operations, AI Practice
Summary
Build and maintain secure, scalable AI infrastructure for Singapore’s government, automating ML pipelines, optimizing LLMs, and enforcing compliance standards.
Software Engineer, AI Infrastructure & Operations, AI Practice at GovTech.
About the role
Join the AI Practice team to define architectural standards and best practices for deploying artificial intelligence across the Singapore government. You will work as a technical consultant and engineer, helping various agencies transition from experimental models to production-ready, secure, and efficient systems.
Key facts Location: Singapore Engagement: Full-time Team: AI Infrastructure & Operations
What you'll do
- Create and manage automated CI/CD pipelines for machine learning, establishing government-wide standards for testing, versioning, and deployment.
- Optimize deep learning models and LLMs using techniques like quantization, pruning, and distillation while evaluating inference engines such as vLLM, TensorRT-LLM, and Triton Inference Server.
- Design scalable multi-tenant infrastructure for training and inference, utilizing GPU orchestration tools like Kubernetes, Volcano, Run:ai, and NVIDIA GPU Operator.
- Build observability frameworks to monitor model health, data drift, GPU telemetry, and system latency.
- Integrate security and compliance guardrails into deployment pipelines to meet IM8 and CSA standards.
- Provide technical consulting to agencies, conducting architecture reviews and creating reusable playbooks to foster self-sufficiency.
Requirements
- Bachelor or Master degree in Computer Science, Computer Engineering, or a related technical field.
- Proven experience in DevOps, MLOps, Cloud Infrastructure, or Platform Engineering with a focus on production AI/ML workloads.
- Proficiency in Python and shell scripting.
- Hands‑on expertise with Docker and Kubernetes in multi-tenant production environments.
- Experience designing MLOps patterns with tools like Kubeflow, MLflow, or BentoML.
- Advanced skills in Infrastructure-as-Code using Terraform or Ansible.
- Experience with GitOps workflows like ArgoCD and platform engineering.
- Deep understanding of GPU hardware, performance tuning, and LLM serving stacks.
- Expertise in observability tools such as Prometheus, Grafana, OpenTelemetry, ELK, or Loki.
- Strong communication skills with the ability to advise both technical teams and agency leadership.
Nice to have
- Postgraduate research in distributed systems, ML systems, or high-performance computing.
- Working knowledge of systems languages such as Go or Rust.
- Certifications in Cloud Architecture or Kubernetes.
- Familiarity with agentic AI infrastructure, RAG pipelines, multi-modal serving, and edge inference.
Skills & tools
- Python, Shell, Go, Rust
- Docker, Kubernetes, Volcano, Run:ai, NVIDIA GPU Operator
- Kubeflow, MLflow, BentoML
- Terraform, Ansible, ArgoCD
- vLLM, TensorRT-LLM, Triton Inference Server, KServe
- Prometheus, Grafana, OpenTelemetry, ELK, Loki, Arize, Langfuse
Practical notes
GovTech provides a total rewards package including wellness programs, leave benefits, and flexible work arrangements. The agency is an equal opportunity employer focused on public sector digital transformation.