Full-Stack Software Engineer — AI-Native Development

SGInnovate; Singapore; On-site; SGD 90,000 - 140,000; Full time

Job summary

AIGP Health is building Anzu, an agentic clinical intelligence platform, in Singapore. We are seeking a capable full-stack engineer to own backend services, data pipelines, and frontend features for health-tech applications.

You will work with Claude Code or similar AI coding agents, deploy on AWS/GCP, and contribute to QA/QC, testing, and robust AI-driven workflows in a fast-paced startup environment.

Qualifications

• 4+ years of full-stack software engineering experience.
• Strong Python skills and ML/data tooling exposure a plus.
• Hands-on cloud experience with AWS and/or GCP.
• Experience with Claude Code or similar AI coding agents in production.

Responsibilities

• Develop backend services, APIs, and data pipelines with a focus on correctness and observability.
• Build frontend components and dashboards for clinical/operational users.
• Lead automated tests and ensure quality gates for health-tech software.
• Collaborate on AI agent workflows and CI-integrated testing.

Skills

Full-stack

Python

Claude Code

AWS/GCP

QA/QC mindset

Testing

AI workflows

Tools

Git

CI/CD

Docker

Kubernetes

PyTorch

Job description

This role is posted on behalf of AIGP, a startup supported by SGInnovate.

Background

AIGP Health is a health-tech company built by doctors and patient advocates who lived the problem of preventable clinical error firsthand. We're building Anzu, an agentic clinical intelligence platform that synthesizes records, checks decisions against safety and evidence, detects care gaps, and automates follow-up. Our mission is health equity and access for all: using clinical AI to reduce preventable harm, give time back to clinicians, and extend high-quality care to more patients across Southeast Asia. We're a small, close-knit team that moves fast, owns problems end-to-end, and stays deeply grounded in the clinicians and patients we build for.

What You'll Do

AI-Native Development Workflow (AIDLC)

• Use Claude Code (and comparable agentic coding tools) daily across the full development lifecycle: spec'ing, scaffolding, implementing, testing, debugging, and documenting.
• Design workflows, prompts, and guardrails (custom commands, skills, subagents, CI hooks) that let AI agents do more of the repetitive and boilerplate work safely and reliably.
• Review AI-generated code and AI-driven test coverage with the same rigor as human-written code — you own correctness, security, and maintainability regardless of who (or what) wrote the first draft.
• Help evolve our team's AIDLC practices: what to automate, what needs a human in the loop, and how to measure whether AI-assisted workflows are actually improving speed and quality.
• Design and build backend services, APIs, and data pipelines in Python, with an emphasis on correctness, performance, and observability.
• Work with ML/data-heavy components (e.g., PyTorch-based services or pipelines) where relevant to product features.
• Build and operate cloud infrastructure on AWS and/or GCP — deployments, storage, data pipelines, and monitoring.
• Own data models and integrations that must hold up to digital health-grade accuracy, security, and auditability requirements.
• Contribute to the development of AI agents, and LLM-based solutions.

Frontend

• Build responsive, accessible, well-tested user interfaces that make complex data and workflows easy to use for clinical, research, or operational end users.
• Collaborate with product and design to turn requirements into clean, maintainable frontend architecture.
• Optimize for performance and usability, including on data-dense screens (dashboards, tables, visualizations).

QA / QC

• Build and maintain automated test suites (unit, integration, end-to-end) as a first-class deliverable of every feature, not an afterthought.
• Define and enforce quality gates appropriate for health-tech software — including traceability, reproducibility, and defect triage.
• Use AI-assisted testing (AI-generated test cases, mutation testing, fuzzing) to increase coverage and catch edge cases faster than manual QA alone.
• Participate in root-cause analysis for production issues and drive fixes upstream into process and tooling, not just the immediate bug.

What We're Looking For

• 4+ years of professional full-stack software engineering experience, comfortable moving between backend, frontend, and testing/QA work.
• Strong Python skills; experience with PyTorch or other ML/data tooling is a plus, especially if you've shipped data- or model-backed features to production.
• Hands‑on experience with cloud platforms — AWS and/or GCP — including deployment, storage, and basic infrastructure‑as‑code practices.
• Real, hands‑on experience using Claude Code or a comparable AI coding agent (Cursor, Github Copilot, etc.) in a production codebase — you can speak concretely to what worked, what didn't, and how you supervised the output.
• A QA/QC mindset: you write tests without being asked, think about failure modes proactively, and treat quality as everyone's job.
• Strong written and oral communication — Clearly document specs, AI prompts, and reviews. Comfortable presenting your work in weekly tech and management meetings.
• Comfort with ambiguity and a fast‑moving startup environment.

Nice to Have

• Experience in health‑tech, biotech, life sciences, or another regulated/high‑stakes domain (e.g., fintech, medtech) — you understand what \"quality bar\" means when the output touches patient or clinical data.
• Familiarity with healthcare data standards and compliance frameworks (e.g., PDPA, HIPAA, GDPR HL7/FHIR, GxP, SOC 2).
• Experience designing custom AI agent workflows and tools
• Prior experience on a small hybrid team where you owned a feature or system end‑to‑end, from design through production support.
• AI‑Native Tooling: Claude Code, custom agent workflows, CI‑integrated AI review and testing