GreenNode (a member of VNG Group) is a leading AI Cloud infrastructure provider in Southeast Asia, running a full-stack Cloud & AI ecosystem: from IaaS (vServer, vNetwork, vStorage) and PaaS (VKS-Kubernetes, vDB, vMonitor) to an advanced AI Stack (Model-as-a-Service, AI Gateway, VectorDB, Agent Platform).
We are expanding our AI product line and looking for a Middle-level Business Analyst who will spend approximately 70% of their time on AI products and 30% on the underlying cloud platform. You will own features end-to-end under the direct mentorship of a Senior BA / Product Owner.
JOB RESPONSIBILITIES
1. AI Product Analysis and Requirements (approximately 70% of the role)
- Own the end-to-end requirement lifecycle for assigned AI features: Model-as-a-Service (model catalog, model deployment, inference endpoints, quota and token-based billing), AI Gateway (routing, rate limiting, API key management, observability), VectorDB / RAG pipelines, and Agent Platform capabilities.
- Work with Product Owners, AI/ML Engineers, Platform Engineers and Design to turn product vision and customer needs into clear, buildable requirements: PRD, user stories, acceptance criteria, use cases.
- Analyze and document AI-specific behaviors that classic BA work does not cover: model lifecycle (register, deploy, serve, monitor, deprecate), token-based usage and pricing logic, prompt and response handling, context windows, streaming responses, latency and throughput expectations, fallback and error behavior.
- Map end-to-end AI user journeys, from a developer obtaining an API key, calling an inference endpoint, to monitoring usage and cost, and identify friction points to feed back into the roadmap.
- Benchmark competitor AI platforms (OpenAI, Bedrock, Vertex AI, Together, Fireworks and others) on features, API design and pricing models; summarize gaps and opportunities for the product team.
- Draft wireframes and prototype UI in Figma for AI console screens (model catalog, playground, endpoint configuration, usage dashboards) to align stakeholders quickly before development starts.
2. Cloud Platform Support (approximately 30% of the role)
- Support requirement analysis for the cloud services that AI products depend on: compute (GPU instances), storage, networking, Kubernetes (VKS) and database services.
- Document integration points between AI services and cloud infrastructure: GPU resource allocation, autoscaling behavior, storage for model artifacts, IAM and permission models, quota and billing integration.
- Contribute to cross-product consistency: shared console UX, common API conventions, unified billing and quota concepts across the Cloud and AI portfolios.
3. Process Modeling and Delivery
- Model business and system flows using BPMN 2.0 and UML sequence / activity diagrams, especially multi-service AI flows (request to gateway, to model serving, to logging, to billing).
- Write clear, testable specifications for Engineering and QA; clarify requirements during sprint execution; participate in grooming, planning and demo.
- Coordinate and execute UAT with internal stakeholders and pilot customers; validate not only functional correctness but also AI-specific quality aspects such as response quality, latency, error handling and cost accuracy.
- Produce user-facing documentation: user guides, API usage documentation and release notes for AI product features.
4. Data-Driven Insights
- Read product usage data, inference logs and customer feedback to surface actionable insights: which models are used, where users drop off, which errors occur most frequently, what drives cost.
- Define success metrics for AI features (adoption, token volume, latency P95, error rate, cost per request) and work with the Data team on reporting dashboards.
- Translate findings into prioritized feature proposals and backlog items.
5. Stakeholder Collaboration
- Act as the bridge between Product, AI/ML Engineering, Platform Engineering, Design and customer-facing teams for assigned features.
- Communicate AI concepts clearly to non-technical stakeholders, and business context clearly to engineers.
- Escalate risks and dependencies early; drive alignment on scope and trade-offs together with the Senior BA / Product Owner.
REQUIREMENTS
Education & Experience
- Bachelor's degree in Information Technology, Business Information Systems, Computer Science, Data Science or a related field.
- 2 to 4 years of experience as a Business Analyst or Product Analyst in technology products.
- At least 1 year working on AI/ML, LLM, data platform, or API / developer-facing products. Alternatively, strong and demonstrable hands-on exposure to AI products (side projects, internal tools, AI agent builds) that you can walk through in detail.
- Proven end-to-end feature ownership within an Agile / Scrum team.
AI Product Knowledge (must-have)
- Solid working understanding of LLM fundamentals: what a model, token, context window, embedding and inference endpoint are; the difference between fine-tuning, prompting and RAG.
- Familiar with RAG architecture: chunking, embedding, vector search, retrieval, re-ranking, generation.
- Familiar with AI Agent concepts: tool / function calling, MCP, multi-step reasoning, memory.
- Understand AI serving concerns at a product level: latency versus throughput, batching, GPU utilization, quotas, rate limiting, token-based pricing..
- Hands-on user of AI tools and agents (ChatGPT, Claude, Copilot, Cursor) to accelerate BA work: drafting documents, generating user stories and test cases, summarizing meetings. Prompt engineering or having built task-specific agents is a strong plus.
Cloud Knowledge
- Understand core cloud concepts: IaaS versus PaaS versus SaaS, compute, storage and network primitives, regions and availability zones.
- Basic familiarity with containers and Kubernetes (pod, deployment, service, autoscaling) and how AI workloads run on them.
- Understanding of GPU compute basics (GPU types, allocation, sharing) is a plus.
Technical & Business Analytics Skills
- Proficient in producing structured documentation: PRD, SRS, user stories, acceptance criteria in Gherkin format.
- Hands-on with BPMN 2.0 and UML (sequence, activity, use case), plus modeling and wireframing tools such as Figma, Lucidchart, draw.io, Miro or Mermaid.
- Comfortable in Agile / Scrum environments and able to adapt across different project management tools (Jira, Redmine, Loop, Notion).
- Able to read and analyze product and usage data to support decisions; basic-to-intermediate SQL and BI tools (Metabase, Looker, Power BI) is a plus.
Soft Skills
- Ownership mindset: you follow a feature through to production, not just to handoff.
- Strong curiosity and self-learning ability; the AI space moves fast and you are expected to keep up.
- Clear communication and facilitation; able to ask sharp questions and challenge assumptions respectfully.
- Structured analytical thinking with attention to detail; comfortable with ambiguity in an early-stage product area.
Language: Able to read and write clear, concise technical documentation in English; comfortable working in a mixed English and Vietnamese environment.
Other Advantages (nice-to-have)
- Experience with AI platform and MLOps tooling: vLLM, Triton, Ray, SageMaker, Vertex AI, Bedrock, LangChain / LlamaIndex, Kubeflow, MLflow.
- Experience with vector databases (Milvus, Qdrant, Weaviate, pgvector) at a product or integration level.
- Exposure to AI evaluation concepts: benchmarks, LLM-as-judge, hallucination and response quality measurement.
- Domain experience in Banking and Financial Services, e-commerce, or enterprise SaaS.
