Senior/Principal AI Domain Architect

Quick Facts

• Role: Senior/Principal AI Domain Architect

• Focus: Reference architecture and governance for Agentic AI solutions; integrations with domain systems

Description

You will design, develop, and oversee the architecture of AI solutions and reference architectures for Agentic AI to ensure alignment with IT strategy, regulations, security principles, and business objectives.

You will define and maintain architectural principles, patterns, guardrails, and decisions across human-in-the-loop, autonomy control, observability, auditing, outcome quality, and model risk management.

You will select AI models/platforms/tools (including Azure AI Foundry, GitHub Copilot, Vertex AI Enterprise, LangChain/LiteLLM) and design integrations of AI agents with domain systems using APIs, MCP, and CI/CD pipelines, supporting a move to Spec-Driven/Agentic SDLC while mentoring teams.

Responsibilities

• Architect AI and Agentic AI reference solutions aligned with security, compliance, and business goals

• Define guardrails and architectural decisions for human-in-the-loop, autonomy limits, observability, auditing, quality, and model risk

• Choose AI platforms and tooling considering cost, security, privacy, vendor lock-in, AI FinOps, vendor risk, and compliance (EU AI Act, DORA, KNF, NIS2, OWASP LLM/Agentic)

• Integrate AI agents with domain systems via APIs, MCP, and CI/CD pipelines; transition from classic SDLC to Spec-Driven/Agentic SDLC with knowledge repositories in JIRA, Confluence, GitHub, and M365

• Collaborate with engineering teams and competency centers in an agile model; build standards, pattern libraries, and knowledge repositories; provide technical mentoring for safe AI usage

Requirements

• 10+ years of professional IT/data/solution/enterprise architecture experience

• At least 3 years as a technical leader/senior architect/domain architect for complex AI/ML, data engineering, MLOps, or data processing systems

• Practical experience with Azure or GCP and GenAI/AI/ML tools (Azure AI Foundry, GitHub Copilot, Vertex AI Enterprise, LangChain/LiteLLM)

• Experience with LLMs, RAG, vector databases, and model/agent evaluation; understanding model limitations, token-cost considerations, privacy, security, and model risk

• Ability to design reference architectures and define principles, patterns, guardrails, and architecture decisions for AI/agentic systems with human-in-the-loop, autonomy boundaries, and auditability

• Understanding of regulated-sector requirements and ability to incorporate EU AI Act, DORA, KNF guidelines, NIS2, and OWASP LLM/Agentic into architecture; AI governance/model risk or ISO/IEC 42001 is an advantage

• Ability to work effectively with Governance, Security, DevOps, Data Science, product owners, engineering teams, and vendors; communicate architecture decisions both technically and business-wise

Benefits

• Stable employment under an employment contract and a hybrid work model

• Continuous learning and development support

• Jointly defined success, with tools and flexibility to make a meaningful impact

• Broad benefits package (cafeteria, private medical care with dental, Multisport card, insurance, internal contests, discount and loyalty programs, 16 hours for volunteering “Dzień na U”)

• Alior University for professional and personal development

• Cash reward for successful referrals via the Referral Program

• Work in an organization actively developing modern relationship-based banking

• Values-driven work: responsibility, openness, innovation, and customer orientation