19 hours ago
Bucharest, RomaniaSenior
Responsibilities
- Translate business and product requirements into target architectures and implementation decisions.
- Design agent, RAG, model-routing, integration, API, identity, and human-in-the-loop patterns.
- Define hybrid cloud and on-premises deployment patterns based on residency, latency, security, performance, and cost constraints.
- Evaluate technology choices and document architecture decisions, trade-offs, and non-functional requirements.
- Provide technical assurance throughout delivery and support production-readiness reviews.
- Design deterministic and agentic workflows, multi-agent patterns, durable state, memory, tool calling, human approval, fallback, and exception handling.
- Design model-agnostic and multi-model architectures with routing, gateways, prompt and policy layers, structured outputs, caching, and latency or cost trade-offs.
- Design MCP and API-based tool integration, event-driven architecture, identity delegation, secrets management, auditability, and zero-trust patterns for agents.
- Produce architecture artifacts for cloud, containerized, Kubernetes, private-network, on-premises, and locally hosted inference deployments.
- Support proposals, client workshops, market development, reusable practice assets, coaching, and continuous learning.
Requirements
- 8+ years of experience in solution, enterprise, cloud, or AI architecture.
- Strong knowledge of generative AI, agentic systems, data platforms, integration, and distributed applications.
- Experience designing hybrid cloud and on-premises solutions.
- Ability to communicate architecture decisions to executives, engineers, security teams, and business owners.
- Hands-on architecture experience with at least two agent orchestration approaches, such as LangChain, LangGraph, Microsoft Agent Framework, Semantic Kernel, OpenAI Agents SDK, AutoGen, or CrewAI.
- Strong knowledge of RAG and knowledge architectures, including embedding models, vector and hybrid search, reranking, metadata filtering, semantic layers, knowledge graphs, context management, and retrieval evaluation.
- Practical knowledge of MCP, API-based tool integration, event-driven architecture, identity delegation, secrets management, auditability, and zero-trust patterns for agents.
- Ability to uphold quality, independence, confidentiality, and risk-management requirements.
Benefits
- Hybrid work arrangement indicated by the #LI-Hybrid designation.
Tech Stack
Categories
AI ApplicationsSolutions Engineering
About PwC
PwC is a global network of professional services firms providing assurance, tax, and advisory work for businesses and public-sector clients. It sells project-based and managed consulting, audit, and deals services, and implements and integrates business applications. Formed in 1998 by the merger of Price Waterhouse and Coopers & Lybrand, it is headquartered in London and is one of the Big Four accounting firms.