Site Reliability Engineer

Binance
Singapore
Workplace: RemoteFull timeFunction: DevOps, Cloud & InfrastructureExperience: 2+ yearsSkills: ["Independent problem solving","Research and experimentation","Rapid prototyping","Learning velocity","Judgment about model behavior"]

Design and operate next-generation retrieval pipelines for agentic RAG systems, with end-to-end production reliability. Build adaptive retrieval and retrieve-reflect-refine workflows using embedding models, vector stores, hybrid search, reranking, and multimodal parsing. Develop agent harness runtimes for recovery, sandbox isolation, multi-tenant execution, and retrieval-grounded tool calling. Create evaluation benchmarks and real-world feedback loops to improve agent performance, groundedness, latency, and task success in production.

Loading

Loading job details...

Preparing the role view and application actions.

FursaFursa
Binance
Binance
10 hours ago

Site Reliability Engineer

✓ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI
Last checked: 5 hours agoStatus: Live

Job Summary

Design and operate next-generation retrieval pipelines for agentic RAG systems, with end-to-end production reliability. Build adaptive retrieval and retrieve-reflect-refine workflows using embedding models, vector stores, hybrid search, reranking, and multimodal parsing. Develop agent harness runtimes for recovery, sandbox isolation, multi-tenant execution, and retrieval-grounded tool calling. Create evaluation benchmarks and real-world feedback loops to improve agent performance, groundedness, latency, and task success in production.
Location: Singapore
Workplace: Remote
Employment Type: Full time
Job Function: DevOps, Cloud & Infrastructure
Seniority: Mid level

Key Responsibilities

  • •Design and operate next-generation retrieval pipelines using adaptive, self-correcting, and multi-hop agentic RAG workflows for production systems.
  • •Architect agentic RAG systems with dynamic retrieval control and iterative retrieve-reflect-refine loops, including multi-agent retrieval collaboration.
  • •Collaborate on model-capability-driven harness innovations such as context management, long-term memory, subagent/multi-agent architectures, and self-evolving agents.
  • •Propose and build benchmark datasets and evaluation methodologies to measure and improve agent intelligence (retrieval efficiency, latency, groundedness, task success rate).
  • •Leverage real-world multi-channel user feedback and production task data to run experiments and continuously improve agent and retrieval performance.

Pay and Benefits

Perks:Remote Work

Key Requirements

  • •2-8+ years hands-on experience building LLM, RAG, and AI agent systems in production.
  • •Hands-on RAG and agentic RAG engineering: production retrieval pipelines with embeddings, vector stores (Qdrant, Milvus, Pinecone, Weaviate), hybrid search, reranking, and chunking/text cleaning; implement Agentic RAG patterns like self/corrective RAG and retrieve-reflect-refine loops.
  • •Agent harness engineering experience with Agent Harness runtimes (Pi Agent, AgentScope 2.0 or equivalent), including session recovery, sandbox isolation, middleware/hooks, multi-tenant runtimes, and plan/execute loops.
  • •Deep familiarity with LLM and agent fundamentals including LLM APIs, KV cache, tool use, planning/reasoning, MCP, memory, subagents and multi-agent systems; strong prompt and context engineering.
  • •Ability to analyze ambiguous problems from first principles and drive research from 0 to 1, translating ideas into runnable prototypes with fast experiment iteration loops.
Experience:2+ yearsLLMRAGAI agents
Skills:Independent problem solvingResearch and experimentationRapid prototypingLearning velocityJudgment about model behavior
Tech Stack:LLMRAGAgentic RAGEmbedding modelsBGEOpenAIQdrantMilvusPineconeWeaviateHybrid searchReranking modelsChunkingText cleaningMultimodal parsingSelf-RAGCorrective RAGAdaptive retrievalMulti-hop decompositionRetrieve-reflect-refine loops

Company Brief

Binance
Operates one of the world’s largest cryptocurrency exchanges, offering spot and derivatives trading, a native token (BNB), wallet services, staking, and a broad suite of crypto financial products and infrastructure.
Industry: Trading Platforms
Company Size: Enterprise (1,001+ employees)
Revenue: USD 1B+
Growth: Established Company
Valuation: Unicorn (USD 1B+)
Funding: Bootstrapped
Headquarters: George Town, Cayman Islands
Founded: 2017
WebsiteLinkedIn