Senior AI System Engineer

Description and Requirements Responsibilities 1. Core Architecture & Delivery: Lead backend architecture design, core development, and delivery for AI & Agent projects. On-Device Integration & Optimization: Collaborate with core teams (e.g., Memory & Knowledge, Orchestration, Agent Runtime) to integrate AI modules (e.g., inference, Agent workflow, RAG, memory/context, Agent orchestration) on device. Ensure high-availability packaging and ultra-low-resource optimization. 2. Cross-Platform & Performance: Build cross-platform (Windows/Android) AI SDKs/apps, with a focus on memory, power, and latency optimization for mobile. 3. Engineering Efficiency: Write technical docs and API specs. Build and maintain automated testing and CI/CD pipelines to ensure quality and efficiency in cross-platform delivery. 4. Tech Leadership: Stay current with AI advancements. Mentor junior/mid-level engineers. Qualifications Education & Language Background Education: Bachelor’s or higher in CS, AI, Software Engineering, or related field (985/211 preferred). Good English for reading docs/papers and global communication. Experience: 5-8+ years in software development, with proven 0-to-1 delivery of complex projects. Core Technical Skills Deep Go Mastery: Master Go core syntax and features, with deep understanding of Goroutine scheduling (G-M-P), Channel communication, GC principles & tuning, and memory allocation (TCMalloc & on-device leak prevention). Go + AI/Agent Experience: Hands-on experience building/integrating LLM backends, Agent Runtimes, or RAG systems with Go. Familiar with (or able to quickly ramp up on) Go AI ecosystem (e.g., LangChainGo), skilled in LLM API integration, and able to package locally deployed models as services. Performance & CGO: Proficient with pprof for high-concurrency/low-latency optimization. Experienced with CGO to resolve Go/C++ lib performance bottlenecks (e.g., llama.cpp). AI Domain Knowledge: Familiar with LLM app development, Prompt Engineering, Agent paradigms (e.g., ReAct, Plan-and-Solve), RAG workflows, and on-device DBs. Additional Preferred Skills Java: Solid Java background to enable smooth architecture, code, and microservice integration with existing Java teams. Python: Strong ability to read Python code (e.g., LangChain, LlamaIndex) and refactor core logic into Go. Rust (Plus): Highly valued for on-device optimization, safe memory management, and cross-platform low-level interaction (e.g., C-FFI). Nice-to-Have On-Device/Extreme Optimization (Strong Plus): Experience with on-device (Mobile/PC) background apps, daemons, or cross-platform SDKs. Experience in extreme memory optimization (OOM protection, defragmentation) and CPU/GPU inference efficiency. Familiar with on-device inference engines (e.g., llama.cpp, ONNX Runtime, CoreML, ExecuTorch). High-quality contributions to Go/Rust/AI open-source projects on GitHub. Cloud-native familiarity with Server-side AI or large-scale distributed systems experience. Knowledge of Kubernetes, gRPC, service mesh, and able to adapt to Enterprise AI expansion. Good product sense (AI-friendly design) and HCI understanding for AI products.