Agentic AI Architect

xebiacee; USA; Remote; USD 170.000 - 210.000; Vollzeit

Zusammenfassung

xebiacee seeks a senior architect to lead enterprise AI foundation initiatives. You will shape the target architecture for shared versus data-specific capabilities, drive proof-of-concept work, and deliver secure, reusable services.

You will coordinate with data, technology, risk and business stakeholders while advancing LLM-based, agentic solutions. The role emphasizes architecture decisions, policy-compliant deployment, and hands-on development of scalable AI platforms within EU-based

Qualifikationen

• 7+ years in software, data or AI engineering with leadership
• Hands-on with LLM-based and agentic systems beyond prototype
• Strong Python skills with agent orchestration, API/MCP tools, and data grounding
• Experience with AI evaluation, observability and lifecycle management (AgentOps/LLMOps)
• Understanding of data management, governance, pipelines or data products
• Experience designing for regulated environments and strong English communication

Aufgaben

• Define the target architecture and assess existing AI initiatives for maturity, value, reuse, scalability, cost and control.
• Compare orchestration, API and MCP integration, data grounding, evaluation, observability and AI lifecycle options.
• Recommend components to build, buy, adapt, reuse or retire with cost, speed, skills and vendor trade-offs.
• Design for security, privacy, resilience, performance, auditability, access controls and cost.
• Build and evaluate a PoC including agents, integrations, tests and evaluation pipelines.
• Collaborate with data, technology, risk and business stakeholders; share knowledge via runbooks and records.

Kenntnisse

LLM/Agent architecture

Agent orchestration

APIs / MCP tools

Data grounding & retrieval

Observability & AI lifecycle

English communication

Regulated environments

Tools

Python

Jobbeschreibung

Role overview

Shape the architecture for an enterprise agentic AI foundation, separating capabilities that can be shared across the organization from those tailored to data management and delivery. The work combines platform assessment, architecture decisions and hands‑on proof-of-concept development, with an emphasis on moving experiments toward secure, reusable services.

Responsibilities

• Define the target architecture and assess existing AI initiatives for maturity, business value, reuse, scalability, cost and control needs.
• Compare orchestration, API and MCP integration, enterprise data grounding, evaluation, observability and AI lifecycle options.
• Recommend which components to build, buy, adapt, reuse or retire, including trade‑offs in total cost, delivery speed, skills and vendor dependence.
• Design for security, privacy, resilience, performance, auditability, explainability, access controls, human oversight and cost.
• Build and evaluate a proof of concept, including agents, integrations, tests and evaluation pipelines on representative workloads.
• Work with data, technology, risk and business stakeholders; share knowledge through pairing, walkthroughs, runbooks and decision records.

Requirements

• At least 7 years in software, data or AI engineering, including recent architecture or technical leadership work.
• Hands‑on experience delivering LLM‑based and agentic systems beyond prototype stage.
• Strong Python skills, with practical experience in agent orchestration, API or MCP tools, and retrieval or grounding on enterprise data.
• Experience with AI evaluation, observability and lifecycle management, such as AgentOps or LLMOps.
• Understanding of data management and delivery, including metadata, quality, lineage, governance, pipelines or data products.
• Experience designing for regulated environments, strong stakeholder communication, and very good written and spoken English.

Benefits and work setup

• Work must be based in the European Union, with authorization to work there.
• The engagement includes knowledge transfer so internal teams can operate and extend the resulting capabilities.