AI Agent Engineer

This role sits on Binance's AI Infra team, working at the boundary of frontier model capability and real-world agent deployment — a hybrid research/engineering position covering agentic RAG, context/memory management, and multi-agent coordination.

• Design and operate adaptive, self-correcting Agentic RAG pipelines
• Collaborate with researchers on context management, long-term memory, and multi-agent architectures
• Propose and build benchmarks/evaluation methodology for retrieval and agent quality
• Use real-world user feedback as a research signal to improve agents in production

• 1+ year hands-on production LLM/RAG/agent experience
• Production retrieval pipeline experience (embeddings, vector stores, hybrid search, reranking)
• Agent-harness runtime experience (session recovery, sandboxing, tool calling)
• Deep familiarity with LLM/agent fundamentals (KV cache, tool use, planning, MCP, memory)
• Heavy personal use of coding/agent tools

Hybrid role based in Singapore; remote/work-from-home arrangements vary by team.