The opportunity
Binance is a leading global blockchain ecosystem behind the world’s largest cryptocurrency exchange by trading volume and registered users. We are trusted by 300+ million people in 100+ countries for our industry-leading security, user fund transparency, trading engine speed,…
What you'll do
Own the full LLM pipeline from data preparation to production real case usage.
Design, iterate and optimize prompts (zero-/few-shot, chain-of-thought,: tool-calling, etc.) to maximize model utility and safety across products and languages.
Build and maintain Retrieval-Augmented Generation (RAG) QA/search systems: that connect to multi-source knowledge bases.
Familiar with vLLM/SGLang inference architectures and have proven experience: deploying and operating LLM services on multi‑GPU or cluster environments.
Design, implement and operate multi‑agent LLM architectures (e.g. LangGraph,: CrewAI, AutoGen) including task decomposition, agent orchestration, memory sharing and tool‑calling workflows.
Develop evaluation pipelines (automatic metrics & human feedback) to measure: prompt and model quality, bias, and hallucination rates.
What they're looking for
- Master’s Degree or higher in Computer Science, Data Science or related field..
- At least 2 years of deep-learning/NLP experience, including 1+ year practical: LLM work (SFT, DPO, RAG, quantization, inference optimization, etc.).
- Demonstrated prompt engineering & tuning expertise (few-shot design,: structured prompting, prefix-/p-tuning, reward re-ranking, safety filtering).
- Practical experience building and deploying multi‑agent LLM workflows, with: understanding of agent‑orchestrator patterns, shared memory, long‑horizon planning and guard‑rail design.
- Proficient in both English and Chinese communication for efficient cross team collaboration
