Back to SearchPrincipal Solution Architect — Agentic AI Platform
Find a vacancy that works for you. Send us your CV to receive a personalized offer.
Find me a job
We're looking for a Principal Solution Architect — Agentic AI Platform to join our team in Portugal in a fully remote working mode. In this role, you will own the end-to-end architecture for an enterprise-grade Agent Development Platform — a cloud-native ecosystem enabling teams to define, orchestrate, deploy, and observe AI agents at scale. You will design cross-cutting capabilities across runtime, tool and gateway integration, identity and policy enforcement, observability, and evaluation frameworks. This position requires deep expertise in agentic AI systems, governance-driven architecture, and LLM-based application design to support secure and scalable enterprise adoption.
- Define cross-cutting architecture for the Agentic AI platform, including runtime, orchestration, gateway/tool integration, identity, and governance
- Design architecture for agent-to-agent (A2A) and agent-to-tool protocols, ensuring interoperability and trust models
- Lead architectural decisions around agent orchestration frameworks (LangGraph, CrewAI, AutoGen, or similar) for enterprise-scale deployment
- Architect runtime controls for authorization, policy enforcement, and non-human identity management
- Define observability and evaluation layers for agent behavior, performance, and LLM output quality
- Integrate advanced LLM capabilities (RAG pipelines, model routing, prompt/context engineering) into system architecture
- Present and defend architectural frameworks in governance and Architecture Review Board (ARB) sessions
- Ensure secure, cloud-native architecture for managed agent runtimes, with AWS as the reference environment
- Collaborate with security, DevOps, and platform engineering teams to uphold regulatory and compliance guidelines
- 6+ years of experience in software architecture, with at least 2 years dedicated to LLM-powered or agentic AI systems in production
- Proven experience implementing at least one agent orchestration framework at scale (e.g., LangGraph, CrewAI, AutoGen, Strands)
- Expertise designing agent-to-tool integration patterns (function calling, MCP, A2A, or equivalent)
- Strong knowledge of agent governance models, identity enforcement for non-human actors, and policy control frameworks
- Deep understanding of LLM application architecture including retrieval augmentation (RAG), evaluation, and optimization
- Demonstrated experience with runtime design, gateway integration, observability strategies, and evaluation pipelines
- Experience presenting architecture for governance and ARB or similar formal technical reviews
- Proficiency in cloud-native architecture with AWS as the primary platform
- Hands-on experience with AWS Bedrock AgentCore components (Runtime, Gateway, Registry, Policy, Evaluation)
- Familiarity with policy-as-code frameworks (Cedar, OPA) for agent authorization
- Experience designing observability frameworks specific to multi-agent systems
- Background in regulated, governed enterprise environments
- Understanding of agent identity standards (non-human identity federation, MS Entra Agent ID or equivalent)