The team builds the next generation of intelligent platforms with data + AI at the core, packaging LLM capability into stable, extensible backend services and engineering platforms that intelligent data analysis and AI agent products are built on.
What you will do
- Own architecture, development and iteration of the AI application backend so the system stays efficient, stable and scalable, and AI features make it from prototype to production.
- Build the LLM evaluation and inference platform, and keep improving inference efficiency and service stability.
- Develop, wrap and debug LLM interfaces. Integrate agentic frameworks and MCP / A2A capability deeply into business systems, and handle call exceptions and fault tolerance so interfaces stay stable and low-latency.
- Contribute to prompt engineering — design and refine prompts and interaction techniques such as SSE for real business scenarios.
- Develop and maintain existing AI systems and components across Python, Java and TypeScript: refactoring, bug fixes, performance tuning and compatibility work.
- Run day-to-day platform operations: monitoring and alerting, incident playbooks, and continuous work on performance and security.
What we are looking for
- Strong analytical and problem-solving skills, clear communication and teamwork, genuine interest in AI and the initiative to keep learning.
- Master's degree or above in computer science, mathematics or a related field, or a Bachelor's degree or above from a leading university (985/211), with strong fundamentals.
- Fluent with AI coding tools — Codex, Claude Code, Cursor and the like.
- Strong Python with a good grasp of common data structures and algorithms; Java / Go / Rust / C++ / TypeScript are a plus, along with the ability to maintain a mixed stack.
- Experience with search, recommendation, text mining or knowledge graph algorithms, or 2+ years of backend experience (AI application backend experience preferred).
- Familiar with mainstream backend architecture and RESTful API design principles, comfortable with common design patterns, and disciplined about coding conventions.
Interested?
Send your CV and the role you are applying for to
recruit@zbyte-inc.comContact: MemoryLake Talent Team