Job Description

Senior AI/ML Engineer (Agentic Systems)

• Working Hours: Mon-Fri (Hybrid)
• Location: Pasir Panjang
• Remuneration: Up to $10,000 + AWS

About the Role

We are seeking a skilled and experienced Software Engineer with a genuine enthusiasm for Generative AI to join our GenAI team. In this position, you will play a central role in building out a scalable Agentic Framework — tackling complex LLM orchestration challenges, integrating external tools and services, establishing agent observability, and designing the backend infrastructure that underpins our AI-powered applications. If you enjoy solving intricate orchestration problems and are comfortable leveraging AI-assisted development practices to accelerate delivery, we'd love to hear from you.

Key Responsibilities

• Agentic Orchestration & Development: Architect and implement advanced orchestration layers using Python-based frameworks such as LangGraph and n8n, as well as cloud-native AI services like AWS Bedrock Agents. Expand agent capabilities by developing and integrating new "skills" and third-party API tool calls.
• Advanced RAG & Cloud Integration: Design and optimise Retrieval-Augmented Generation (RAG) pipelines. Work extensively with large language models via the AWS ecosystem (Bedrock) and internal government API gateways, with a strong focus on context window management and embedding strategies.
• Platform & Marketplace Architecture: Build the backend architecture for an AI capabilities marketplace — enabling users to securely publish, version, share, and manage datasets, custom document chatbots, AI agents, and agentic skills.
• Backend & API Engineering: Develop robust, scalable backend services and APIs. Design and maintain relational databases (PostgreSQL) to support application state management and complex AI workflows.
• Vector Search & Data Handling: Manage and tune search and retrieval mechanisms using vector databases and search engines (e.g., OpenSearch) to maximise retrieval accuracy and minimise latency.
• Agent Observability & Evaluation: Instrument AI agents with telemetry to track metrics such as token consumption, response latency, and tool-call outcomes. Develop evaluation mechanisms to assess LLM output quality and reduce hallucinations.
• AI Guardrails & Enterprise Security: Implement security controls around LLM usage, including defences against prompt injection, data masking, and output filtering — ensuring full compliance with government data privacy requirements.
• Data Pipelines & Platform Support: Design and maintain data ingestion and processing pipelines (ETL) to supply and refresh vector databases.
• Full-Stack Collaboration: Partner closely with frontend engineers and UI/UX designers to integrate AI backend processes with React and Next.js frontends.

Qualifications

• Experience: At least 5 years of software engineering experience, with a strong foundation in backend development, API design, and system architecture.
• GenAI/LLM Proficiency: Demonstrated hands-on experience building production applications using AI middleware and orchestration frameworks (e.g., LangChain, LlamaIndex, LangGraph), as well as managed LLM services.
• Cloud Expertise: Solid familiarity with major cloud platforms. Hands-on knowledge of AWS services — particularly Bedrock, Bedrock Agents, and OpenSearch — is strongly preferred; equivalent experience on Azure or GCP will also be considered.
• Database Skills: Strong working knowledge of relational database systems, particularly PostgreSQL, along with a solid understanding of vector database concepts.
• Modern Development Practices: Practical understanding of frontend frameworks (React/Next.js); ability to apply AI-assisted coding tools to accelerate development while upholding rigorous standards for code quality, security, and debuggability.