Key Responsibilities
AI System Ownership & Delivery
- Own or co-own AI initiatives from problem definition, technical strategy, architecture design, to production deployment.
- Make informed decisions on model selection, RAG architecture, agent orchestration, and system trade-offs under real-world constraints.
- Design and optimize LLM inference pipelines, embeddings, and system performance for reliability and scalability.
Product-Oriented AI Engineering
- Collaborate closely with business, product, and leadership teams to translate ambiguous requirements into AI-driven solutions.
- Design structured, reusable Prompt Engineering and agent workflows to ensure controllability, robustness, and exploitability.
- Evaluate build vs. buy decisions across models, agent platforms, and infrastructure.
Multi-Agent Systems & Reasoning
- Design and orchestrate multi-agent systems using frameworks such as LangChain, LangGraph, MCP, or equivalent.
- Implement reasoning paradigms including ReAct, Chain-of-Thought (CoT), Tree-of-Thought (ToT).
- Assess and integrate agent platforms (e.g., Coze, Dify, FastGPT) when appropriate.
RAG & Knowledge Infrastructure
- Design and iterate on Retrieval-Augmented Generation (RAG) architectures.
- Build and optimize knowledge systems using vector databases such as Milvus, FAISS, or Chroma.
- Continuously improve retrieval quality, context grounding, and reasoning accuracy.
- Track emerging AI trends in model alignment, agent systems, and multimodal AI.
- Contribute to internal standards, documentation, prototypes, and technical decision frameworks.
- Mentor engineers or collaborate with external partners when needed.
QUALIFICATIONS
Required
- Bachelor's degree or above in Computer Science, AI, or a related field.
- Previous experience founding, co-founding, or being an early technical member of an AI startup, or leading AI products in a startup environment.
- Proven delivery of at least one end-to-end AI product (LLM / RAG / Agent-based system) in production.
- Strong hands-on experience with LLMs, Prompt Engineering, RAG pipelines, and agent frameworks.
- Solid understanding of ReAct-style agent workflows and multi-agent system design.
- Experience making technical decisions under uncertainty, cost, and time constraints.
Nice to Have
- Experience with LoRA / QLoRA, model alignment, or inference optimization.
- Exposure to AI product commercialization, user feedback loops, or go-to-market iteration.
- Open-source contributions, technical writing, or public speaking.
- Strong cross-functional communication and leadership skills.
