About the role
We are seeking a highly skilled and visionary AI Architect to design and lead the architecture of cutting-edge AI systems involving agent orchestration, LLM integration, and enterprise-scale deployment. You will play a pivotal role in selecting the right models, frameworks, and strategies to enable intelligent multi-agent ecosystems, integrating them seamlessly into existing enterprise environments.
Responsibilities:
- Design the overall AI system architecture, focusing on agent orchestration, system integration, and end-to-end scalability.
- Evaluate and select appropriate LLMs, orchestration frameworks, tools, and APIs aligned with use case complexity and enterprise constraints.
- Architect knowledge management systems using RAG (Retrieval-Augmented Generation) and hybrid search patterns.
- Enable plan-and-execute strategies and multi-agent coordination for complex task automation.
- Ensure secure, cost-optimised design through API gateway integration, token and context-window budgeting, and resource utilisation planning.
- Drive Gen AI threat modelling, safety tier design, and defences against adversarial prompts.
- Collaborate with cross-functional teams to ensure seamless A2A (agent-to-agent) and tool-based orchestration.
- Lead proofs-of-concept, technical assessments, and production readiness for agent-based applications.
Requirements:
- Strong understanding of architecture patterns including knowledge management (RAG, hybrid search), agent design, and enterprise AI.
- Hands-on expertise with LLM orchestration frameworks such as LangGraph, CrewAI, AutoGen, Vertex AI, and AI Refinery.
- Proficiency in tools-as-agents, function calling, API integration, model context protocols (e. g., MCP), and A2A communication.
- Experience with Gen AI safety design, including threat modelling and prompt protection strategies.
- Knowledge of API gateway design, performance optimisation, cost-aware system architecture, and context budgeting.
- Experience in working across cloud-native and hybrid environments with secure enterprise integration patterns.
Additional Information:
- The candidate should have a minimum of 5 years of experience in Large Language Models.
- A 15-year full-time education is required.