AI Agent Engineer

Location: Asia; remote

Company

Binance operates a global blockchain ecosystem offering cryptocurrency trading, financial services, payments, institutional products, and Web3 features.

What you will do

• Design and operate adaptive, self-correcting, multi-hop retrieval pipelines and Agentic RAG systems.
• Develop context management, long-term memory, subagent, multi-agent, and self-evolving agent architectures.
• Build harness-domain and RAG-domain benchmarks, datasets, annotation strategies, and evaluation methodologies.
• Measure and improve retrieval efficiency, latency, groundedness, task success rate, and agent intelligence.
• Use user feedback and real-world task data to run experiments and improve production retrieval and agent performance.

Requirements

• 2–8+ years of hands-on experience with LLM, RAG, and AI agent systems in production.
• Experience building end-to-end retrieval pipelines with embedding models, vector stores, hybrid search, reranking, chunking, text cleaning, and multimodal data parsing.
• Experience with Agentic RAG patterns such as Self-RAG, Corrective RAG, adaptive retrieval, multi-hop decomposition, and retrieve-reflect-refine loops.
• Experience with agent orchestration runtimes, session recovery, sandbox isolation, middleware and hooks, multi-tenant runtimes, plan/execute loops, and retrieval-grounded tool calling.
• Strong knowledge of LLM APIs, KV cache, agent loops, tool use, reasoning, planning, MCP, memory, subagents, multi-agent systems, prompt engineering, and context engineering.
• Ability to independently research ambiguous problems, rapidly build prototypes, use AI-assisted development workflows, and iterate experiments quickly.

Culture & Benefits

• Remote work arrangement for the business team.
• Collaboration with international talent in a flat, user-centric organization.
• Autonomy on fast-paced and innovative projects.
• Career growth and continuous learning opportunities.
• Competitive salary and company benefits.