About this role
The Backend Engineer, AI (Agent Systems) position at bjakcareer is located in China and suits backend software professionals focused on AI inference, orchestration layers, and production scalability. The role involves designing and operating systems that bridge machine learning models and end-user applications.
Role Overview
The Backend Engineer, AI (Agent Systems) takes ownership of the inference and orchestration layer driving artificial intelligence interactions within the product. Based in China, the professional in this role develops production systems that convert model capabilities into stable, low-latency, and observable APIs for mobile and desktop clients.
Key Responsibilities
Requirements
Skills and Tools
About the Company
ACTAI develops proactive applications designed to bring intelligence to daily conversations, errands, organization, and workflows with minimal prompting. The organization focuses on long-running workflows with high reliability, persistent context, and real-world task completion to reduce hallucinations. Operating as a small, hands-on team with high talent density, the group makes collective decisions and moves rapidly to build products for users globally.
How to Apply
Candidates should apply directly on the official company careers page.
Skills and tech stack
Questions candidates ask
What is the location of the Backend Engineer, AI (Agent Systems) role?
The position is located in China.
What does the backend engineer own in this role?
The engineer owns the inference and orchestration layer that powers every AI interaction in the product.
What technologies and tools are part of the tech stack?
The tech stack includes Python, Node.js, PyTorch, OpenAI, Anthropic, open-source LLMs, SQL and NoSQL databases, Kubernetes, and Docker.
How many interview rounds can candidates expect?
Candidates can expect 3, but no more than 4 interviews.
What is the primary mission of the company?
The company aims to build proactive applications for anyone worldwide, bringing intelligence to conversations, organization, errands, and workflows with minimal to no prompting.