Tóm tắt công việc
- Design and own end-to-end architecture for AI-powered systems, including multi-agent platforms, production LLM pipelines, and intelligent automation workflows.
- Define system blueprints for AI agent ecosystems: agent orchestration, memory management, tool integration, reasoning loops, and human-in-the-loop fallback.
- Evaluate and select technology stacks (frameworks, infrastructure, models, data stores) aligned with business goals and cost constraints.
- Act as the technical bridge between business stakeholders, clients, and engineering teams - translate product vision into architectural decisions and present solutions clearly to non-technical audiences.
- Lead technical discovery with enterprise clients: gather requirements, propose architecture, estimate effort, and write technical proposals.
- Collaborate with Middle/Senior
AI Engineers who own implementation details - your focus is structural soundness, scalability, maintainability, and system integration.
- Establish architectural standards, design review processes, and best practices for AI system development.
- Stay current with the AI agent landscape (agentic frameworks, LLM routing, observability, safety/guardrails) and recommend when to adopt vs. wait.
- Support pre-sales and bidding activities: architecture demos, PoC design, technical presentations.
Must-Have:
- Bachelor's degree or higher in Computer Science, AI/ML,
Software Engineering, or a related field.
- 5+ years in software engineering, with at least 2 years designing production AI/ML systems.
- Strong system design skills: distributed systems, microservices, event-driven architecture, API design, and data pipelines.
- Solid understanding of AI agent architecture - core components: agent orchestrator, memory (short-term/long-term), tool-use layer, planning/reasoning module, observability, and guardrails.
- Experience architecting LLM-based applications: RAG pipelines, agentic workflows (LangChain, CrewAI, AutoGen, or similar), prompt management, and model routing.
- Ability to communicate complex technical concepts to executives, clients, and cross-functional teams - in Vietnamese and English.
- Hands-on experience with cloud platforms (Sunteco Cloud, AWS, or GCP), containerization (Docker, Kubernetes), and CI/CD for AI systems.
- Strong understanding of trade-offs: cost vs. latency vs. accuracy, open-source vs. managed services, synchronous vs. async agent patterns.
Good-to-Have:
- Experience in pre-sales, technical writing, or client advisory roles.
- Familiarity with MLOps/LLMOps pipelines (MLflow, LangFuse, Braintrust, or similar observability tools).
- Knowledge of vector databases (Milvus, Qdrant, Chroma) and hybrid search strategies.
- Experience with multi-tenant AI platform design.
- A track record of published architectures, talks, or technical blog posts.
- Understanding of AI safety, bias detection, and responsible AI practices.