Product AI Application Backend Engineer
About the Role
We are using AI to fundamentally reshape our core business infrastructure. As an AI Application Backend Engineer, you will be a core builder of AI engineering capabilities — deeply integrating large language models with financial business scenarios to construct the backend AI service layer powering trading intelligence, smart risk control, compliance monitoring, and user services.
Responsibilities
AI Service Layer Architecture & Development: Design and develop core AI backend services, including model gateway (Model Gateway), Agent scheduling layer, and streaming inference services. Build high-concurrency, low-latency, horizontally scalable microservice architectures that meet production SLA requirements for AI-powered services.
Agent & Tool-Calling System Development: Develop Agent workflows and tool-calling (Tool Use / Function Calling) systems on the Openclaw platform. Define tool registration and integration standards for internal systems (trading engine, risk systems, data platforms), ensuring call-level security and result integrity for financial scenarios.
Financial Scenario AI Service Delivery: Own the development and iteration of AI backend services for core scenarios including intelligent customer service, risk signal detection, compliance analysis, and personalized services. Collaborate with algorithm teams to deliver end-to-end model integration, effectiveness validation, and performance tuning.
Data & Storage Layer: Design and maintain data infrastructure required by AI applications, including session memory and long-term memory storage, knowledge base management and incremental update mechanisms — ensuring data consistency and retrieval quality.
Observability & Evaluation: Establish full-chain LLM application tracing, prompt version management, and evaluation frameworks. Design automated LLM-as-Judge pipelines to continuously monitor output quality, hallucination rates, and business KPIs.
Requirements
Programming Fundamentals: Proficiency in Golang or Java; solid understanding of data structures, algorithms, and system design; familiarity with RESTful API and gRPC interface design.
Backend Engineering Experience: 5+ years of backend development experience with production-level expertise in microservice frameworks (Spring Boot / go-zero), message queues (Kafka / RabbitMQ), caching (Redis), and relational databases (PostgreSQL / MySQL).
LLM/Agent Engineering in Production: 1+ year of hands-on production experience building LLM applications or AI Agents, with demonstrated AI skills (Agent development, LLM toolchain integration); able to independently deliver AI feature modules end-to-end.
Prompt Engineering & Model Understanding: Deep understanding of LLM capabilities and limitations (hallucination, context length, reasoning); hands-on experience with Prompt Engineering, Few-shot and Chain-of-Thought design, with the ability to independently craft and iterate prompt strategies for complex business scenarios.