Senior AI/ML Engineer

About the Singapore AI Center of Excellence (COE) The Singapore AI Center of Excellence (COE) is a dedicated engineering and R&D hub focused on Enterprise and Sovereign AI. Our mission is to help organizations move AI workflows cleanly into production through two main paths: creating reusable software architectures that solve common industry problems, and contributing code directly to upstream open-source projects to fix enterprise gaps in system deployment, runtime tuning, and platform management. Basing our engineering team in Singapore creates a close feedback loop between customers, partners, and core product teams. This direct connection keeps our development roadmaps relevant, speeds up solution delivery, and strengthens our ability to co-innovate across the region. Role Overview The Senior AI/ML Engineer is a highly technical, hands-on role at the intersection of Enterprise AI and client-facing architecture. As part of our Customer Engineering function, you will write production-grade solution blueprints alongside strategic customers and technology partners. Your mission is to solve immediate, high-stakes operational bottlenecks in the APAC region by delivering repeatable, extensible, and open-sourced AI Quickstarts that show the industry how to solve complex challenges. In this team, career growth and seniority are defined purely by your technical competence, architectural depth, and ability to deliver end-to-end solutions autonomously in highly ambiguous environments. There is no expectation of team management, project coordination, or formal talent mentorship; your progression is driven entirely by engineering impact. What You Will Do: • Enterprise-Minded Blueprinting: Design and build comprehensive, production-ready architecture blueprints and reference codebases. These blueprints must naturally take into account critical enterprise requirements—including systems-level hardening, infrastructure scalability, and network isolation boundaries—without you needing to perform the last-mile hands-on production deployment yourself. • Open-Source AI Quickstarts: Package repeatable, extensible technical architectures as open-sourced AI Quickstarts to solve complex, real-world industry problems and accelerate ecosystem adoption. • Co-Development & Integration: Collaborate with external engineering teams (such as semiconductor partners, regional AI programs, and software vendors) to validate joint-architecture blueprints, ensuring stable integrations across the system stack. • Benchmarking & Evaluation: Design and integrate automated testing, evaluation harnesses, and system-level telemetry into blueprints to monitor and measure performance metrics like latency, throughput, cost, and model quality. • Regulated & Secure Design: Proactively incorporate robust data privacy standards, secure network perimeters, and localized hosting considerations into all solution blueprints to satisfy compliance and risk management requirements in highly regulated environments. What You Will Bring: Academic & Professional Experience • Education: Bachelor’s or Master's degree in Computer Science, Computer Engineering, or a related quantitative field. • Software Engineering Foundations: Excellent understanding of software engineering fundamentals, including clean code principles, test-driven development (TDD), CI/CD automation pipelines, and version control (Git) workflows. • Communication: Clear verbal and written communication skills in English, with the ability to articulate complex technical architectures to other engineers and technical stakeholders. Core Technical Stack • Programming Languages: Exceptional, hands-on proficiency in Python (specifically for machine learning and systems programming). Solid familiarity with lower-level system languages such as Go or C/C++ is highly preferred. • Deep Learning & ML Libraries: Strong familiarity with PyTorch and core NLP/vision ecosystems (e.g., Hugging Face Transformers , datasets, and tokenizers). • Cloud-Native Frameworks: Strong practical experience deploying containerized applications on Kubernetes or production-grade enterprise container orchestration platforms. • Base MLOps Knowledge: Conceptual and hands-on understanding of model serving lifecycles, data ingestion steps, and automated packaging. Target Knowledge Domains & Growth Areas We are building a multi-disciplinary engineering squad. Candidates are expected to bring experience in some of the following domains, and will have the opportunity to continuously develop their skills across all of them as they grow in seniority: • Platform & Pipeline Engineering: Experience with distributed computing frameworks and cluster schedulers (such as Ray ), workflow orchestration (such as MLflow or Kubeflow ), distributed unstructured data parsing tools (such as Docling ), and vector database structures. • Generative AI & Agentic Architectures: Familiarity with LLM orchestration engines (such as LangChain , LlamaIndex , or LangGraph ), agentic workflows (AgentOps), tool-calling protocols (such as Model Context Protocol ) and model fine-tuning (PEFT/SFT). • Hardware Heterogeneity & Runtime Optimization: Understanding of model serving runtimes (such as vLLM ), optimizing engines for latency and throughput (e.g., KV cache offloading, chunked prefill), and running benchmarks across diverse hardware setups including CUDA, ROCm, and emerging GPU/NPU architectures. • Regulated Deployments: Designing isolated, air-gapped container networks, secure registry services, and local inference environments to meet strict data residency, privacy, and compliance policies. Nice-to-Have: Industry Domain Experience While not strictly required, experience applying AI/ML architectures to solve challenges in the following regulated industries is a strong advantage: • Financial Services (FSI): Familiarity with risk-assessment models, fraud detection patterns, or compliance requirements under regulatory framewor…