Backend Engineer AI Agent
Summary
Backend engineer in Singapore building scalable backend services and AI Agent-powered products for the client's AI team. Day to day involves Golang services, MySQL/Redis/message queues/Elasticsearch, and integrating LLM capabilities such as agent workflows, tool calling, and RAG into product features.
Backend Engineer
to help build backend systems and
AI Agent-powered products
. You will work closely with product, data, and engineering teams to turn AI capabilities into practical product features and scalable backend services. This role is ideal for a backend engineer who enjoys solving complex engineering problems and is interested in the practical application of
LLMs, AI Agents, and AI-powered workflows
.
Job Responsibilities Design and develop scalable backend systems and services for AI-powered products. Contribute to system architecture, technical design, and feature development. Build backend infrastructure for
AI Agent applications
, including agent workflows, tool calling, task orchestration, and external integrations. Integrate LLM capabilities into practical product workflows with a focus on reliability and scalability. Work with cross-functional teams to translate business requirements into technical solutions. Explore AI technologies and AI-assisted development tools to improve engineering efficiency.
Job Requirements Bachelor's degree or above in Computer Science, Software Engineering, or a related field. 3+ years of backend development experience
with strong programming fundamentals. Strong proficiency in
Golang
; experience with Java/Spring Boot is a plus. Experience with
MySQL, Redis, message queues, Elasticsearch
, or similar technologies. Experience with system architecture, API design, distributed systems, or middleware. Strong interest in
LLMs, AI Agents, and AI-powered applications
. Experience with
LLM APIs, agent frameworks, tool calling, RAG, or workflow orchestration
is a plus. Strong problem-solving and communication skills. Proficiency in
English and Chinese
is preferred.