Full Time Lead Engineer - Engineer Agentic Platform, AI Platform, xCloud Jobs, in HTX (Home Team Science & Technology Agency) - Maukerja

What The Role Is

HTX is Singapore's Science and Technology agency that brings together diverse scientific and engineering capabilities to develop transformative, operationally ready solutions for public safety. As a statutory board under the Ministry of Home Affairs, HTX works at the forefront of science and technology to empower the Home Team with cutting-edge capabilities. Guided by our mission to amplify, augment and accelerate the Home Team's advantage, we are committed to keeping Singapore the safest place on planet Earth.

xCloud is dedicated to elevating the enterprise experience through cutting-edge cloud capabilities, including:

• Sustainable enterprise data centres
• Hybrid cloud platforms
• Advanced cloud security measures
• Integrated Development, Security, and Operations (DevSecOps)
• Innovative enterprise Software-as-a-Service (SaaS) solutions
• AI-powered enterprise applications
  
  

The jobholder will be part of the AI Platform team, building the common capabilities used by teams across HTX and the Home Team to develop and operate secure agentic AI systems.

You will design, build and operate the engineering components that enable AI agents to securely reason, use tools, access enterprise services and execute multi-step workflows.

You will work on capabilities including agent runtimes, Model Context Protocol (MCP), tool gateways, skills, secure sandboxes, durable execution, human approvals, evaluations, CI/CD and agent observability.

What You Will Be Working On

• Develop and maintain services used to execute AI agents.
• Implement agent execution loops integrating models, tools, context and application state.
• Integrate agent harnesses and development frameworks into the platform.
• Implement structured model inputs and outputs, tool calling and response validation.
• Implement retries, timeouts, cancellation and error handling around model and tool calls.
• Develop APIs and SDKs through which applications invoke agent capabilities.
• Implement resource limits governing execution duration, model calls, token consumption and tool calls.
• Troubleshoot complex agent execution failures across application, model and infrastructure layers.
• Develop and operate MCP servers, clients and gateways.
• Implement APIs and connectors that enable agents to securely interact with approved enterprise systems.
• Define machine-readable tool schemas with clear inputs, outputs and error contracts.
• Implement authentication and authorisation for tool calls. Build tool discovery and registration capabilities.
• Build secure environments for executing model-generated code and other untrusted agent actions.
• Implement workload isolation using approved container, Kubernetes, microVM or sandbox technologies.
• Apply CPU, memory, storage and execution-time constraints to sandbox workloads.
• Implement filesystem and network restrictions.
• Build mechanisms for safely injecting temporary and scoped credentials when required.
• Ensure sandbox workloads cannot access platform credentials or unauthorised services.
• Build capabilities for long-running and multi-step agent execution.
• Implement workflow state persistence, retry, timeout, scheduling and recovery mechanisms.
  
  

What We Are Looking For

• 1–5+ years of relevant experience, with appointment and scope as Engineer or Lead Engineer commensurate with technical capability and experience.
• Strong software-engineering fundamentals.
• Strong proficiency in at least one programming language such as Python, Go, TypeScript/JavaScript or Java.
• Experience developing APIs, backend services or distributed applications.
• Hands-on experience with Docker and Kubernetes.
• Experience with modern CI/CD practices and tools such as GitLab CI.
• Experience with Helm and Infrastructure as Code or GitOps practices.
• Understanding of modern LLM concepts including model APIs, prompting, structured outputs, tool/function calling and context management.
• Understanding of agentic AI concepts and multi-step model execution.
• Familiarity with MCP or comparable tool-integration approaches.
• Experience implementing automated unit and integration testing.
• Knowledge of application and infrastructure observability using technologies such as Prometheus, Grafana or OpenTelemetry.
• Understanding of API security, authentication and authorisation.
• Understanding of DevSecOps and secure software-development practices.
• Strong troubleshooting ability across software and infrastructure layers.
• Ability to learn new frameworks and technologies rapidly.
• Good written and verbal communication skills.
  
  

All new hires are appointed on a two-year contract in the first instance and will be assessed and considered for permanent tenure over time, based on performance.

As part of the shortlisting process for this role, you may be required to complete a medical declaration and/or undergo further assessment.

All applicants will be updated on the status of their applications within 4 weeks upon closing of the advertisement.