AI/ML工程师

Build and deploy LLM-powered applications (RAG, tool-using agents, workflow automation); Design and implement multi-agent orchestration systems (agent roles, routing, memory, tool-calling, evaluation); Develop solutions using frameworks such as CrewAI, LangChain, LangGraph, LlamaIndex; Fine-tune models (e.g., LoRA/QLoRA, PEFT, SFT) and optimize inference quality/cost/latency; Integrate LLMs with internal/external tools and APIs (function calling, structured outputs, observability); Create evaluation pipelines (prompt tests, regression suites, grounding checks, hallucination mitigation); Collaborate with product/engineering to deliver features quickly with strong ownership.

Minimum 2 years of professional experience in AI/ML or LLM application development; Strong experience with Python and modern AI/ML tooling; Hands-on experience with CrewAI and/or LangChain, plus orchestration patterns (agents, tools, memory); Experience with LangGraph for graph-based agent workflows or equivalent; Experience with LlamaIndex (or similar) for RAG pipelines, indexing, retrieval, and embeddings; Proven ability to fine-tune an LLM (open-source models preferred) and measure improvements; Practical knowledge of Claude tooling / Claude Code and modern AI developer workflows; Familiarity with emerging/open ecosystem tools (e.g., OpenClaw or similar agent/tool frameworks); Strong problem-solving skills and a builder mentality you ship, iterate, and improve; Experience deploying LLM systems to production (Docker, CI/CD, Kubernetes, serverless).

If your profile is shortlisted, you will receive an email from our company within 24–48 hours. Please check your inbox (and spam/junk folder) during this period.