Senior AI Engineer (Multi-Agent Systems)

We are representing a high-growth technology innovator in Singapore that is currently scaling its AI Capabilities Hub. As the "Agentic Era" takes hold, our client is moving beyond simple chatbots to complex, autonomous ecosystems.

We are looking for a Senior AI Engineer who is not just a consumer of LLMs, but an architect of Multi-Agent Systems (MAS). You will be instrumental in designing systems where multiple specialised agents collaborate, reason, and execute complex workflows to solve real-world industrial challenges.

Key ResponsibilitiesArchitect & Orchestrate: Design and deploy scalable multi-agent architectures using frameworks such as LangGraph, CrewAI, or AutoGen to handle complex, non-linear tasks.System Reasoning: Implement advanced reasoning patterns (e.g., ReAct, Reflexion, Chain-of-Thought) to enhance agent decision-making and error-correction capabilities.Production-Grade AI: Lead the transition from experimental notebooks to robust, production-ready microservices, ensuring low latency and high reliability in agent communications.Evaluation & Observability: Develop custom "Evals" and monitoring frameworks to track agentic drift, cost-efficiency, and task accuracy.Cross-Functional Leadership: Act as a technical bridge between product owners and engineering teams to identify high-value "agentic" opportunities within the existing product suite.

Technical RequirementsExperience: Minimum of 5 years in Software Engineering, with at least 2 years focused specifically on Generative AI and Agentic Workflows.Core Stack: Expert-level Python (asyncio, FastAPI) and experience with Vector Databases (e.g., Pinecone, Weaviate, or Milvus).Agentic Mastery: Proven track record of building and deploying Multi-Agent Systems. You should be comfortable discussing the trade-offs between hierarchical vs. joint-collaboration agent structures.LLM Proficiency: Deep understanding of prompt engineering, fine-tuning (LoRA/QLoRA), and RAG (Retrieval-Augmented Generation) optimisations.Infrastructure: Hands-on experience with containerisation (Docker/Kubernetes) and cloud AI workbenches (AWS Bedrock, Azure AI Studio, or GCP Vertex AI).

Preferred AttributesContributions to open-source agent frameworks or AI research.Experience in a fast-paced "Zero-to-One" environment.A "Product Mindset"—you care as much about the user outcome as you do about the model's perplexity.

Why Apply?This is a chance to work on the "bleeding edge" of AI in 2026, moving away from simple wrappers and into the future of autonomous software. Our client offers a highly collaborative culture, a dedicated R&D budget for experimentation, and a clear pathway to Lead/Principal roles as the team expands.