Full Stack Engineer
About the job
About the Full Stack Engineer role
Key of Responsibilities:
The Engineer will be embedded within the Client AI Programme team and will:
- Design, build, test, deploy, and operate production AI applications across for clients
- Write high-quality full-stack software across TypeScript, NestJS, React, Postgres, and AWS
- Integrate LLM and multimodal APIs in production, including prompting, tool/function calling, streaming, voice interaction, retrieval, and memory patterns
- Build evaluation pipelines for AI quality, safety, regression testing, and human review
- Improve observability across AI products, including traces, logs, metrics, user feedback loops, and model behaviour monitoring
- Build and harden prototypes of emerging AI capabilities into maintainable product or platform components
- Collaborate with product managers, designers, data scientists, governance colleagues, GovTech Central, and client stakeholders
- Senior/Staff engineers will additionally be expected to lead architecture decisions, mentor engineers, set engineering standards across AI product squads, and represent the team in technical discussions with GovTech Central and client stakeholders
Requirements:
- Strong software engineering fundamentals: system design, testing, debugging, CI/CD, observability, security, and maintainable code
- Demonstrated ability to build production-grade full-stack applications and services
- Hands-on experience (or strong working knowledge) of LLM application development
- Clear thinking about AI system quality: evals, regression tests, failure modes, monitoring, and human review
- Ability to communicate clearly with both engineers and non-technical stakeholders
Preferred Qualifications:
- Experience with TypeScript, NextJS, React, Vite, Postgres, AWS, Langfuse, or OpenTelemetry
- Experience integrating AI services such as OpenAI, Anthropic, or other LLM/multimodal APIs
- Experience with RAG, tool calling, agentic workflows, voice interfaces, or AI evaluation frameworks
- Familiarity with education, public-sector systems, or high-trust digital services
