Kỹ sư Thuật toán at Deepcoin

1. Role Highlights:

• Participate in building AI workflow orchestration and automation platforms from scratch or scaling existing ones (RAG, tool invocation, Agent collaboration, asynchronous orchestration).
• Implement Diffy or similar "shadow traffic replay + response diff comparison" solutions for model/version regression testing and gradual rollout.
• Collaborate with multiple business lines (IM, customer service, marketing automation, data processing, etc.) to deliver real production-level closed-loop solutions.

2. Key Responsibilities:

• AI/LLM Workflow Orchestration: Design and implement multi-step reasoning, Agent collaboration, tool invocation (Tool-Calling/Function-Calling), asynchronous task queues, and compensation mechanisms.
• RAG Optimization: Build and optimize Retrieval-Augmented Generation pipelines including data ingestion, chunking, vectorization, recall/reranking, context compression, caching, and cost reduction.
• Evaluation & Quality Assurance: Establish automated evaluation and alignment systems (benchmark sets, Ragas/G-Eval/custom metrics), integrate A/B testing, and implement online monitoring.
• Engineering & Observability: Develop model/prompt version management, feature/data versioning, experiment tracking (MLflow/W&B), and audit logs.
• Platform Integration: Expose workflows via API/SDK/microservices; integrate with business backends (Go/PHP/Node), queues (Kafka/RabbitMQ), storage (Postgres/Redis/object storage), and vector databases (Milvus/Qdrant/pgvector).

3. Must-Have Qualifications:

• 3+ years backend or data/platform engineering experience with 1-2 years hands-on LLM/generative AI projects.
• Proficiency in LLM application engineering: prompt engineering, function/tool calling, dialogue state management, memory, structured output, alignment and evaluation.
• Experience with workflow frameworks: LangChain/LangGraph, LlamaIndex, Temporal/Prefect/Airflow, or custom DAG/state machine implementations.
• End-to-end RAG experience: data cleaning → vectorization → recall → reranking → evaluation.
• Familiarity with Diffy or equivalent traffic replay/diff comparison solutions.
• Strong engineering fundamentals: Docker, CI/CD, Git, observability tools (OpenTelemetry/Prometheus/Grafana).
• Proficiency in Go/Python/TypeScript with ability to write reliable services and tests.

4. Preferred Qualifications:

• Deep Diffy implementation experience.
• LLMOps/evaluation platform experience (Arize Phoenix, Evidently, PromptLayer).
• Agent framework implementation (LangGraph, autogen/crewAI).
• Security/compliance knowledge (data masking, GDPR).
• Domain experience in IM/customer service/marketing automation.
• Cost optimization techniques for LLM deployments.

Competitive compensation, career growth opportunities, collaborative team environment, full remote work flexibility.