We are looking for a hands-on Software Engineer with strong AI engineering capabilities who can build production AI systems and work directly with the teams using them. This role spans full-stack development, LLM applications, evaluation, observability, deployment, and customer-facing technical delivery. Accessibility experience is not required, but you should be willing to develop expertise in digital accessibility and WCAG standards.
Responsibility
- Design, develop, and maintain high-quality features and services for government product teams, ensuring accessibility and performance.
- Develop full-stack features across frontend, backend, APIs, data, and AI systems.
- Design and build production-grade applied AI-powered applications and developer tools.
- Utilise CI/CD pipelines to ensure seamless, reliable, and high-quality deployments.
- Develop and maintain technical documentation, including API specs, troubleshooting guides, and system architecture diagrams.
- Translate customer feedback and production issues into reusable product improvements.
- Work alongside Product Management to provide L1/L2 support to customers we serve.
- Work directly with government teams on discovery, demos, Proof-of-Concepts, architecture discussions, onboarding, integrations, and troubleshooting.
- Gain knowledge debugging and remediating WCAG accessibility issues on production websites
Requirements
- Proven experience in designing and building scalable, reliable, and maintainable software systems.
- Strong proficiency in modern software development (e.g., React.js, TypeScript, Node.js) and a deep understanding of browser architecture.
- Build LLM workflows using RAG, tool calling, structured outputs, agents, and human-in-the-loop patterns.
- Build evaluation pipelines using golden datasets, regression tests, deterministic evaluators, and LLM-as-a-judge.
- Implement observability, monitoring, guardrails, fallbacks, and validation for reliable AI systems.
- Balance model quality, latency, reliability, maintainability, and cost.
- Maintain automated tests, CI/CD pipelines, and technical documentation.
- Hands-on experience with quality engineering practices, including automated testing (unit, integration, E2E) and test automation frameworks.
- Solid grasp of CI/CD, infrastructure concepts, and agile engineering practices.
- Strong troubleshooting and debugging skills, with an ability to analyze complex system behaviors and logs.
- Excellent communication skills to explain technical trade-offs and quality standards to diverse stakeholders.
Nice to Haves
- Docker, Infrastructure as a Code and Bash scripting knowledge
- Passion for building inclusive technology and a willingness to apply or develop expertise in WCAG 2.2 standards and assistive technologies.
- Willingness to learn knowledge of digital accessibility, including digital accessibility regulations & Web Content Accessibility Guidelines (WCAG) 2.2 Level A and AA requirements, learn how to use screen readers such as VoiceOver, NVDA and TalkBack.
- Experience writing code that meets accessibility standards such as the Web Content Accessibility Guidelines (WCAG).
- Candidates with more than 7 years of relevant experience in software engineering may be considered for senior positions.
Benefits
By submitting your resume/CV, you consent and agree to allow the information provided to be used and processed by or on behalf of Xtremax Pte Ltd for purposes related to your registration of interest in current or future employment with us and for the processing of your application for employment.
You also represent to us that you have obtained the consent of your referees when you disclose to us their personal data for the purpose of conducting reference checks.
The personal data held by us relating to your application will be kept strictly confidential and in accordance with the PDPA. You may also refer to our Privacy Policy for more details here:
We regret to inform you that should you not consent to providing the necessary data required for us to process your application, your application will be considered void.
