AI Software Engineer at POSable Sdn Bhd
This Full time on site position offers great opportunities for career growth. Applied AI, Full-Stack Development & QA Automation Reporting to: Chief Technology Officer Work location: In person Eligibility: Malaysian citizens only
About the Role
We are looking for an AI Software Engineer with strong development fundamentals and a genuine passion for building practical, production-ready AI systems. You will work directly with the CTO across two key platforms: A transaction-critical platform supporting sales, payments, inventory, customers, reporting, and enterprise integrations. An AI-powered application that enables users to work with documents, databases, business data, and external information through natural language. This is a hands-on mid-level role covering applied AI, full-stack development, system modernisation, and automated quality assurance.
What You Will Work On
- AI assistants that understand user intent and complete multi-step tasks.
- Context engineering, prompt design, conversation state, long-term memory, and personalisation.
- Agent skills, tool calling, workflow orchestration, and AI automation.
- Retrieval-augmented generation, document ingestion, embeddings, semantic search, reranking, and vector databases.
- Natural-language querying of databases, spreadsheets, documents, and business information.
- AI evaluation, regression testing, observability, latency, reliability, security, and cost optimisation.
- Transactional features involving sales, payments, inventory, reconciliation, reporting, and enterprise integrations.
- Frontend applications, backend services, REST APIs, databases, and cloud deployment.
- Refactoring and modernising existing systems without introducing regressions.
- QA and Test Automation Build and maintain automated unit, integration, API, regression, and end-to-end tests.
- Develop browser automation using Playwright and related testing tools.
- Automate critical user journeys across transactional and AI-powered workflows.
- Create repeatable evaluation suites for AI routing, retrieval, memory, tool usage, response quality and failure handling. realistic truncated description to maintain format response limitations.